# cloud cost management

Cloud cost management is the process of managing and reducing an organization's overall cloud spending, using monitoring, forecasting, governance, and optimization practices.

This is one page of public article previews, not the complete archive. Follow Next page to continue. Summaries are not the original full articles.

## Finding Amazon S3 cost savings with natural language using AWS Quick and S3 Tables

DevFeed: [Finding Amazon S3 cost savings with natural language using AWS Quick and S3 Tables](<https://devfeed.tech/articles/finding-amazon-s3-cost-savings-with-natural-language-using-aws-quick-and-s3-tables-143610.md>)

Original publisher: [Read original article](<https://aws.amazon.com/blogs/storage/finding-amazon-s3-cost-savings-with-natural-language-using-aws-quick-and-s3-tables/>)

Author: Santosh Setty

Published: 2026-11-18T23:59:46Z

Content type: tutorial

Language: en

Sources: [AWS Storage Blog](<https://devfeed.tech/sources/aws-storage-blog.md>)

Topics: [Amazon S3 Tables](<https://devfeed.tech/topics/amazon-s3-tables.md>), [Amazon S3](<https://devfeed.tech/topics/amazon-s3.md>), [cloud cost management](<https://devfeed.tech/topics/cloud-cost-management.md>), [Amazon CloudWatch](<https://devfeed.tech/topics/amazon-cloudwatch.md>), [Artificial Intelligence](<https://devfeed.tech/topics/ai.md>), [text2sql](<https://devfeed.tech/topics/text2sql.md>)

Tags: [advanced-300](<https://devfeed.tech/tags/advanced-300.md>), [amazon](<https://devfeed.tech/tags/amazon.md>), [amazon-s3](<https://devfeed.tech/tags/amazon-s3.md>), [amazon-s3-tables](<https://devfeed.tech/tags/amazon-s3-tables.md>), [amazon-simple-storage-service-amazon-s3](<https://devfeed.tech/tags/amazon-simple-storage-service-amazon-s3.md>), [aws](<https://devfeed.tech/tags/aws.md>), [aws-cloud-storage](<https://devfeed.tech/tags/aws-cloud-storage.md>), [aws-quick](<https://devfeed.tech/tags/aws-quick.md>), [cost](<https://devfeed.tech/tags/cost.md>), [cost-savings](<https://devfeed.tech/tags/cost-savings.md>), [costs](<https://devfeed.tech/tags/costs.md>), [data](<https://devfeed.tech/tags/data.md>), [growth](<https://devfeed.tech/tags/growth.md>), [logs](<https://devfeed.tech/tags/logs.md>), [metrics](<https://devfeed.tech/tags/metrics.md>), [natural-language](<https://devfeed.tech/tags/natural-language.md>), [overhead](<https://devfeed.tech/tags/overhead.md>), [query](<https://devfeed.tech/tags/query.md>), [questions](<https://devfeed.tech/tags/questions.md>), [replication](<https://devfeed.tech/tags/replication.md>), [request](<https://devfeed.tech/tags/request.md>), [s3](<https://devfeed.tech/tags/s3.md>), [s3-tables](<https://devfeed.tech/tags/s3-tables.md>), [tables](<https://devfeed.tech/tags/tables.md>), [technical-how-to](<https://devfeed.tech/tags/technical-how-to.md>)

### AI overview

Combining S3 Storage Lens metrics with object-level server access logs in S3 Tables lets AWS Quick answer natural-language questions about storage costs across buckets. Storage Lens provides bytes, object counts, and request metrics by storage class and prefix; access logs show object reads but not storage size. The walkthrough exports daily metrics to the aws-s3 table bucket, sends logs through CloudWatch Logs to aws-cloudwatch, then connects both to Quick. Example analyses identify cold prefixes, incomplete multipart uploads, noncurrent versions, cross-Region replication and reads, and rising GET traffic. Setup requires Quick Enterprise, and initial data may take 1-2 days to become useful; logs include sensitive requester and object details. Quick's findings are recommendations: verify the SQL or query in Athena and confirm retention needs before applying permanent expiration rules.

### Source excerpt

Across thousands of S3 buckets, the data you need to cut storage costs sits in different places. Learn how to bring Amazon S3 Storage Lens metrics and server access logs together in S3 Tables and query them with AWS Quick. Find cold data, wasted storage, replication overhead, and request growth by asking questions in natural language.

## How Airbnb uses cost per task to operate coding agents on Amazon Bedrock

DevFeed: [How Airbnb uses cost per task to operate coding agents on Amazon Bedrock](<https://devfeed.tech/articles/how-airbnb-uses-cost-per-task-to-operate-coding-agents-on-amazon-bedrock-141280.md>)

Original publisher: [Read original article](<https://aws.amazon.com/blogs/machine-learning/how-airbnb-uses-cost-per-task-to-operate-coding-agents-on-amazon-bedrock/>)

Author: Alan Yao

Published: 2026-11-18T21:48:19Z

Content type: article

Language: en

Sources: [Artificial Intelligence](<https://devfeed.tech/sources/artificial-intelligence.md>)

Topics: [Amazon Bedrock](<https://devfeed.tech/topics/amazon-bedrock.md>), [LLM observability](<https://devfeed.tech/topics/llm-observability.md>), [cloud cost management](<https://devfeed.tech/topics/cloud-cost-management.md>), [Amazon CloudWatch](<https://devfeed.tech/topics/amazon-cloudwatch.md>), [agentic-coding](<https://devfeed.tech/topics/agentic-coding.md>), [AI-assisted coding](<https://devfeed.tech/topics/ai-assisted-coding.md>), [rate-limiting](<https://devfeed.tech/topics/rate-limiting.md>)

Tags: [advanced-300](<https://devfeed.tech/tags/advanced-300.md>), [airbnb](<https://devfeed.tech/tags/airbnb.md>), [amazon-bedrock](<https://devfeed.tech/tags/amazon-bedrock.md>), [coding-agents](<https://devfeed.tech/tags/coding-agents.md>), [cost](<https://devfeed.tech/tags/cost.md>), [customer-solutions](<https://devfeed.tech/tags/customer-solutions.md>), [evals](<https://devfeed.tech/tags/evals.md>), [how-to](<https://devfeed.tech/tags/how-to.md>), [latency](<https://devfeed.tech/tags/latency.md>), [model-selection](<https://devfeed.tech/tags/model-selection.md>), [observability](<https://devfeed.tech/tags/observability.md>), [production](<https://devfeed.tech/tags/production.md>), [quotas](<https://devfeed.tech/tags/quotas.md>), [spending](<https://devfeed.tech/tags/spending.md>), [tasks](<https://devfeed.tech/tags/tasks.md>)

### AI overview

Airbnb combines task-based evaluations with Amazon Bedrock production metrics to select and revisit coding-agent models. It scores models and reasoning-effort settings on representative engineering tasks, pricing all uncached input, cache reads, cache writes, and output tokens across a task; these results guide defaults and effort ceilings. Production can diverge: one model scored better at similar eval cost, but its users later spent several times more because requests cost about three times as much and users sent about three times as many requests. A newer model later scored higher at roughly one sixth the task cost, and heavy users' spend fell to about a seventh with similar request volume. For operations, Airbnb tracks latency, token counts, derived output speed, throttles, and estimated quota usage. Quota admission reserves uncached input, cache writes, and max_tokens, so lowering an oversized max_tokens can improve admission headroom without changing final quota consumption for unchanged output. Gateway records, IAM tags, inference profiles, and invocation logs support aggregate spend attribution; logs require access controls and retention settings.

### Source excerpt

Airbnb runs evals on real engineering tasks and uses Amazon Bedrock observability to choose models for its coding agents and control production costs. This post shows how cost per task guides model selection, why production spending can differ from eval results, and how to monitor latency, manage quotas, and attribute spend.

## Keep skyrocketing AI costs in check

DevFeed: [Keep skyrocketing AI costs in check](<https://devfeed.tech/articles/keep-skyrocketing-ai-costs-in-check-141277.md>)

Original publisher: [Read original article](<https://www.cio.com/article/4243812/keep-skyrocketing-ai-costs-in-check.html>)

Author: Snowflake and Google Cloud

Published: 2026-11-18T21:31:59Z

Content type: article

Language: en

Sources: [CIO](<https://devfeed.tech/sources/cio.md>)

Topics: [cost-optimization](<https://devfeed.tech/topics/cost-optimization.md>), [cloud cost management](<https://devfeed.tech/topics/cloud-cost-management.md>), [finops](<https://devfeed.tech/topics/finops.md>), [AI Infrastructure](<https://devfeed.tech/topics/ai-infrastructure.md>), [Frontier AI](<https://devfeed.tech/topics/frontier-ai.md>)

Tags: [agentic-ai](<https://devfeed.tech/tags/agentic-ai.md>), [ai](<https://devfeed.tech/tags/ai.md>), [ai-coding](<https://devfeed.tech/tags/ai-coding.md>), [ai-costs](<https://devfeed.tech/tags/ai-costs.md>), [ai-economics](<https://devfeed.tech/tags/ai-economics.md>), [ai-infrastructure](<https://devfeed.tech/tags/ai-infrastructure.md>), [ai-usage](<https://devfeed.tech/tags/ai-usage.md>), [artificial-intelligence](<https://devfeed.tech/tags/artificial-intelligence.md>), [budget](<https://devfeed.tech/tags/budget.md>), [budgeting](<https://devfeed.tech/tags/budgeting.md>), [cloud](<https://devfeed.tech/tags/cloud.md>), [coding](<https://devfeed.tech/tags/coding.md>), [controls](<https://devfeed.tech/tags/controls.md>), [cost](<https://devfeed.tech/tags/cost.md>), [costs](<https://devfeed.tech/tags/costs.md>), [edition-us](<https://devfeed.tech/tags/edition-us.md>), [finops](<https://devfeed.tech/tags/finops.md>), [google](<https://devfeed.tech/tags/google.md>), [google-cloud](<https://devfeed.tech/tags/google-cloud.md>), [infrastructure](<https://devfeed.tech/tags/infrastructure.md>), [language-en](<https://devfeed.tech/tags/language-en.md>), [limits](<https://devfeed.tech/tags/limits.md>), [llm](<https://devfeed.tech/tags/llm.md>), [storytype-brandpost](<https://devfeed.tech/tags/storytype-brandpost.md>)

### AI overview

AI costs can rise quickly as agentic systems consume more tokens and AI workloads depend on model choice and prompt iteration. The text describes three Snowflake approaches, developed with Google Cloud: granular spend views, quotas and group budgets with programmable enforcement; model flexibility, including routing tasks among approved open or proprietary models using cost and performance trade-offs; and an agent harness that loads tools on demand, bundles calls, compacts outputs and supplies context to reduce calls, turns and tokens. These controls aim to improve cost visibility and predictability while supporting AI adoption. The text does not quantify savings.

### Source excerpt

The rapid rise of AI has introduced another challenge for IT leaders: How to effectively manage surging compute and infrastructure costs with an eye towards optimizing spend and achieving demonstrable ROI. Expensive AI infrastructure, rising development costs, and token-based pricing models are helping to drive up the AI price tag, often to levels that are higher than expected. Agentic AI takes a greater toll on costs as models use between 5 and 30 times more tokens per task than standard chatbots. By 2028, Gartner predicts AI coding costs will overtake the average developer's salary due to surging LLM token consumption and the shift to consumption-based licensing models. A FinOps Foundation survey identified managing the cost and use of tokens in SaaS-model AI as the top challenge facing practitioners today. "Agents have blown out the cost of everything," says Rachel Roundy, principal product marketing manager for AI at Snowflake. "Companies are at a point where they are weighing the pros and cons of how much value they are getting out of the agentic toolset versus how much they are paying for it. They are looking for ways to find cost efficiencies." A trifecta for keeping costs in check Compared to cloud compute workloads, AI tasks stack up costs more quickly depending on the model and prompt iteration cycle, creating budgeting and forecasting challenges. Snowflake, in tandem with partner Google Cloud, delivers capabilities and services to help optimize AI economics and bring predictability to deployments. Among them Budget and cost controls. Snowflake establishes transparency and trust on the cost front in a variety of ways. Highly granular views of daily AI spend provide visibility into adoption patterns and AI usage across business units so quick adjustments can be made. User quotas and group budgeting capabilities allow organizations to set spending limits for defined groups of users and resources while triggering automated and programmable cost enforcement. "

## From Prompts to Management Reports: Snowflake HTML Reports with CoCo

DevFeed: [From Prompts to Management Reports: Snowflake HTML Reports with CoCo](<https://devfeed.tech/articles/from-prompts-to-management-reports-snowflake-html-reports-with-coco-140791.md>)

Original publisher: [Read original article](<https://medium.com/snowflake/from-prompts-to-management-reports-snowflake-html-reports-with-coco-2febecfc5e8c?source=rss----34b6daafc07---4>)

Author: Omid Karami

Published: 2026-11-18T15:01:04Z

Content type: article

Language: en

Sources: [Snowflake Builders](<https://devfeed.tech/sources/snowflake-builders.md>)

Topics: [finops](<https://devfeed.tech/topics/finops.md>), [cloud cost management](<https://devfeed.tech/topics/cloud-cost-management.md>), [AI Chat](<https://devfeed.tech/topics/ai-chat.md>), [data analyst](<https://devfeed.tech/topics/data-analyst.md>)

Tags: [2026](<https://devfeed.tech/tags/2026.md>), [ai](<https://devfeed.tech/tags/ai.md>), [ai-assisted](<https://devfeed.tech/tags/ai-assisted.md>), [artificial-intelligence](<https://devfeed.tech/tags/artificial-intelligence.md>), [attribution](<https://devfeed.tech/tags/attribution.md>), [coco](<https://devfeed.tech/tags/coco.md>), [cost](<https://devfeed.tech/tags/cost.md>), [data](<https://devfeed.tech/tags/data.md>), [data-engineer](<https://devfeed.tech/tags/data-engineer.md>), [data-engineering](<https://devfeed.tech/tags/data-engineering.md>), [data-platform](<https://devfeed.tech/tags/data-platform.md>), [engineer](<https://devfeed.tech/tags/engineer.md>), [engineering](<https://devfeed.tech/tags/engineering.md>), [experiment](<https://devfeed.tech/tags/experiment.md>), [explore](<https://devfeed.tech/tags/explore.md>), [exploring](<https://devfeed.tech/tags/exploring.md>), [finops](<https://devfeed.tech/tags/finops.md>), [html](<https://devfeed.tech/tags/html.md>), [management](<https://devfeed.tech/tags/management.md>), [prompts](<https://devfeed.tech/tags/prompts.md>), [reports](<https://devfeed.tech/tags/reports.md>), [snowflake](<https://devfeed.tech/tags/snowflake.md>), [snowflake-coco](<https://devfeed.tech/tags/snowflake-coco.md>)

### AI overview

Snowflake CoCo and HTML Reports can turn Account Usage queries into shareable cost reports, but the report's usefulness depends on validating its data and SQL. In this case study, the report covered service, warehouse, query, storage, and CoCo usage; observed consumption spanned only November 9-10 of the requested 30-day window, and database storage returned no rows. Service-level CoCo metering was 5.1573 credits versus 5.3893 at the model level; warehouse figures also differed, with causes unverified. The reviewed SQL aggregated CoCo credits and tokens separately by model before joining, addressing a potential join-fanout concern without proving full financial reconciliation. HTML Reports are published snapshots: refreshing workspace data does not update a published artifact, which must be republished. During Public Preview, scheduled refresh and delivery were unavailable. Because viewers may access embedded data without source-table privileges, authors should review results, sensitive identifiers, and audience permissions before publishing. The report supports investigation, not automatic cost optimization or verified savings.

### Source excerpt

Exploring Snowflake's new HTML Reports capability through a practical cost consumption report -- and how AI-assisted reporting can support FinOps, cost visibility, and better platform decisions. AI-generated image created using ChatGPT by OpenAI.TL;DR Snowflake HTML Reports, introduced in Public Preview in November 2026, let practitioners use Snowflake CoCo to turn query results into shareable reports. This case study demonstrates a Snowflake Cost Consumption Report generated using the bundled cost-intelligence skill. It includes the original prompt, executed SQL, report screenshots, and validation findings. The experiment illustrates how AI-assisted reporting can support management decisions and FinOps, while highlighting the importance of data completeness, SQL verification, and transparent limitations. 1. From Data to Decisions: The Management Reporting Challenge "Where are our Snowflake credits going?" Answering this question often requires examining several Account Usage views. Service metering, warehouse consumption, query attribution, storage, and AI-assisted development activity each provide different perspectives. For Data Platform Teams, the challenge extends beyond retrieving metrics. Technical results must be organized into reports that stakeholders can understand, question, and use to prioritize investigations. I chose to explore whether CoCo and HTML Reports could bridge the gap between technical data exploration and management reporting. As a data engineer working with Snowflake, I see value in making platform consumption understandable beyond engineering teams. Cost consumption was a practical starting point because Snowflake already provides detailed Account Usage data. This allowed me to evaluate two things together: the AI-assisted reporting experience and the accuracy of the generated results. The objective was not to demonstrate automatic cost optimization or claim financial savings. It was to investigate whether an AI-assisted workflow could pro

## Your Data Stack Should Live in the Cloud | Blog | Fivetran

DevFeed: [Your Data Stack Should Live in the Cloud | Blog | Fivetran](<https://devfeed.tech/articles/your-data-stack-should-live-in-the-cloud-blog-fivetran-140204.md>)

Original publisher: [Read original article](<https://wf.fivetran.com/blog/cloud-is-future>)

Author: Charles Wang

Published: 2026-11-18T02:47:08Z

Content type: opinion

Language: en

Sources: [Fivetran Blog](<https://devfeed.tech/sources/fivetran-blog.md>)

Topics: [cloud-migration](<https://devfeed.tech/topics/cloud-migration.md>), [Managed Services](<https://devfeed.tech/topics/managed-services.md>), [self-hosted databases vs managed services](<https://devfeed.tech/topics/self-hosted-databases-vs-managed-services.md>), [cloud-infrastructure](<https://devfeed.tech/topics/cloud-infrastructure.md>), [data mobility](<https://devfeed.tech/topics/data-mobility.md>), [cloud cost management](<https://devfeed.tech/topics/cloud-cost-management.md>)

Tags: [cloud](<https://devfeed.tech/tags/cloud.md>), [cloud-based](<https://devfeed.tech/tags/cloud-based.md>), [cloud-data-warehouse](<https://devfeed.tech/tags/cloud-data-warehouse.md>), [cloud-native](<https://devfeed.tech/tags/cloud-native.md>), [connectors](<https://devfeed.tech/tags/connectors.md>), [data](<https://devfeed.tech/tags/data.md>), [data-infrastructure](<https://devfeed.tech/tags/data-infrastructure.md>), [data-integration](<https://devfeed.tech/tags/data-integration.md>), [data-pipeline](<https://devfeed.tech/tags/data-pipeline.md>), [data-pipelines](<https://devfeed.tech/tags/data-pipelines.md>), [data-stack](<https://devfeed.tech/tags/data-stack.md>), [downtime](<https://devfeed.tech/tags/downtime.md>), [engineers](<https://devfeed.tech/tags/engineers.md>), [enterprise-resource-planning](<https://devfeed.tech/tags/enterprise-resource-planning.md>), [fivetran](<https://devfeed.tech/tags/fivetran.md>), [fortune-500](<https://devfeed.tech/tags/fortune-500.md>), [free](<https://devfeed.tech/tags/free.md>), [managed](<https://devfeed.tech/tags/managed.md>), [managed-services](<https://devfeed.tech/tags/managed-services.md>), [services](<https://devfeed.tech/tags/services.md>)

### AI overview

The article argues that organizations can respond to growing data volumes and pressure for faster decisions by moving their data stack to managed cloud services. It says engineers otherwise bear the burden of building and maintaining many connectors and coordinating data flows; on-premises infrastructure adds hardware design and scaling demands, and integration and analytics may compete for computing resources. The proposed approach is a managed data pipeline paired with a cloud data warehouse, which the article says can provide elasticity, performance, and lower cost while preserving engineering effort, reducing operating costs, and limiting downtime.

### Source excerpt

For most organizations, on-premises is over. The future? Managed services in the cloud.

## Your AI coding agent could cost 5x more depending on the harness, benchmark finds

DevFeed: [Your AI coding agent could cost 5x more depending on the harness, benchmark finds](<https://devfeed.tech/articles/your-ai-coding-agent-could-cost-5x-more-depending-on-the-harness-benchmark-finds-139336.md>)

Original publisher: [Read original article](<https://thenewstack.io/coding-agent-harness-costs/>)

Author: Amanda Caswell

Published: 2026-11-17T22:13:52Z

Content type: news

Language: en

Sources: [The New Stack](<https://devfeed.tech/sources/the-new-stack.md>)

Topics: [Agent Harness](<https://devfeed.tech/topics/agent-harness.md>), [Benchmark](<https://devfeed.tech/topics/benchmark.md>), [cloud cost management](<https://devfeed.tech/topics/cloud-cost-management.md>), [DeepSeek Harness](<https://devfeed.tech/topics/dsh.md>)

Tags: [22](<https://devfeed.tech/tags/22.md>), [agent](<https://devfeed.tech/tags/agent.md>), [agent-harness](<https://devfeed.tech/tags/agent-harness.md>), [ai](<https://devfeed.tech/tags/ai.md>), [ai-agents](<https://devfeed.tech/tags/ai-agents.md>), [ai-coding](<https://devfeed.tech/tags/ai-coding.md>), [ai-coding-agent](<https://devfeed.tech/tags/ai-coding-agent.md>), [ai-model](<https://devfeed.tech/tags/ai-model.md>), [benchmark](<https://devfeed.tech/tags/benchmark.md>), [can](<https://devfeed.tech/tags/can.md>), [coding](<https://devfeed.tech/tags/coding.md>), [coding-agent](<https://devfeed.tech/tags/coding-agent.md>), [cost](<https://devfeed.tech/tags/cost.md>), [developer-tools](<https://devfeed.tech/tags/developer-tools.md>), [finops](<https://devfeed.tech/tags/finops.md>), [first](<https://devfeed.tech/tags/first.md>), [harness](<https://devfeed.tech/tags/harness.md>), [it-s](<https://devfeed.tech/tags/it-s.md>), [model](<https://devfeed.tech/tags/model.md>), [more](<https://devfeed.tech/tags/more.md>), [run](<https://devfeed.tech/tags/run.md>), [times](<https://devfeed.tech/tags/times.md>), [up](<https://devfeed.tech/tags/up.md>)

### AI overview

A benchmark of 22 coding-agent harness configurations across five open-weight models found that harness choice can change costs by as much as fivefold for the same model, while rankings often shift sharply when the model changes. The model explained most cost variation (92.8%), but no harness was consistently cheapest across models. Nearly 65% of estimated inference spending went to failed or timed-out runs; only 35.5% went to successful runs. Step count was the only tested factor with a statistically significant correlation to cost per successful task (0.51). The five best-value harnesses relied heavily on shell commands and rarely used subagents or context summarization, but the study did not isolate those design choices, so it cannot establish that they caused lower costs. Estimates used Baseten's published rates.

### Source excerpt

The same AI model can cost up to five times more to run depending on the coding agent harness it's The post Your AI coding agent could cost 5x more depending on the harness, benchmark finds appeared first on The New Stack.

## Microsoft's new coding model promises near-Opus performance at 5% of the cost

DevFeed: [Microsoft's new coding model promises near-Opus performance at 5% of the cost](<https://devfeed.tech/articles/microsoft-s-new-coding-model-promises-near-opus-performance-at-5-of-the-cost-138569.md>)

Original publisher: [Read original article](<https://thenewstack.io/microsoft-mai-code-2/>)

Author: Frederic Lardinois

Published: 2026-11-17T19:14:48Z

Content type: news

Language: en

Sources: [The New Stack](<https://devfeed.tech/sources/the-new-stack.md>)

Topics: [LLM evaluation / benchmarking](<https://devfeed.tech/topics/llm-evaluation-benchmarking.md>), [ai evaluation](<https://devfeed.tech/topics/ai-evaluation.md>), [post-training](<https://devfeed.tech/topics/post-training.md>), [model-deployment](<https://devfeed.tech/topics/model-deployment.md>), [cloud cost management](<https://devfeed.tech/topics/cloud-cost-management.md>), [Microsoft](<https://devfeed.tech/topics/microsoft.md>)

Tags: [agentic](<https://devfeed.tech/tags/agentic.md>), [ai](<https://devfeed.tech/tags/ai.md>), [ai-models](<https://devfeed.tech/tags/ai-models.md>), [anthropic](<https://devfeed.tech/tags/anthropic.md>), [benchmarks](<https://devfeed.tech/tags/benchmarks.md>), [claude](<https://devfeed.tech/tags/claude.md>), [claude-haiku-5-5](<https://devfeed.tech/tags/claude-haiku-5-5.md>), [claude-opus](<https://devfeed.tech/tags/claude-opus.md>), [claude-opus-5-5](<https://devfeed.tech/tags/claude-opus-5-5.md>), [code](<https://devfeed.tech/tags/code.md>), [coding](<https://devfeed.tech/tags/coding.md>), [conference](<https://devfeed.tech/tags/conference.md>), [cost](<https://devfeed.tech/tags/cost.md>), [developer-tools](<https://devfeed.tech/tags/developer-tools.md>), [ignite](<https://devfeed.tech/tags/ignite.md>), [microsoft](<https://devfeed.tech/tags/microsoft.md>), [model](<https://devfeed.tech/tags/model.md>), [models](<https://devfeed.tech/tags/models.md>), [opus](<https://devfeed.tech/tags/opus.md>), [performance](<https://devfeed.tech/tags/performance.md>), [promises](<https://devfeed.tech/tags/promises.md>)

### AI overview

Microsoft says MAI Code 2 reaches 98% of GPT-6.1 Sol's performance at 40% of its cost and 96% of Claude Opus 5.5's at 5% of its cost. Code 2 Flash is reported to match Claude Haiku 5.5 and GPT-6 Luna on DeepSWE 1.1, Terminal-Bench 2.1, and SWE Atlas at 20% and 65% of their respective costs. These figures are Microsoft's claims; the article notes OpenAI had released newer models, which weakens the comparison. Both models are tuned for routine coding work rather than fundamental science or long-horizon agentic tasks. MAI Code 2 and Flash are available in GitHub Copilot; Flash is also in limited preview on Microsoft Foundry, with additional providers expected to add support.

### Source excerpt

At its Ignite conference in San Francisco on Tuesday, Microsoft announced a number of new AI models, including two new The post Microsoft's new coding model promises near-Opus performance at 5% of the cost appeared first on The New Stack.

## Cloud repatriation is on the rise -- but tricky questions remain

DevFeed: [Cloud repatriation is on the rise -- but tricky questions remain](<https://devfeed.tech/articles/cloud-repatriation-is-on-the-rise-but-tricky-questions-remain-137002.md>)

Original publisher: [Read original article](<https://www.cio.com/article/4245967/cloud-repatriation-on-the-rise-but-tricky-questions-remain.html>)

Author: Stephanie Overby

Published: 2026-11-17T10:01:00Z

Content type: news

Language: en

Sources: [CIO](<https://devfeed.tech/sources/cio.md>)

Topics: [cloud-migration](<https://devfeed.tech/topics/cloud-migration.md>), [Cloud](<https://devfeed.tech/topics/cloud.md>), [Local AI](<https://devfeed.tech/topics/local-ai.md>), [cloud cost management](<https://devfeed.tech/topics/cloud-cost-management.md>), [digital sovereignty](<https://devfeed.tech/topics/digital-sovereignty.md>)

Tags: [ai](<https://devfeed.tech/tags/ai.md>), [ai-workloads](<https://devfeed.tech/tags/ai-workloads.md>), [cios](<https://devfeed.tech/tags/cios.md>), [cloud](<https://devfeed.tech/tags/cloud.md>), [cloud-computing](<https://devfeed.tech/tags/cloud-computing.md>), [cloud-computing-cloud-management-data-center-hybrid-cloud-private-cloud](<https://devfeed.tech/tags/cloud-computing-cloud-management-data-center-hybrid-cloud-private-cloud.md>), [cloud-management](<https://devfeed.tech/tags/cloud-management.md>), [cloud-migration](<https://devfeed.tech/tags/cloud-migration.md>), [cost](<https://devfeed.tech/tags/cost.md>), [data-center](<https://devfeed.tech/tags/data-center.md>), [databases](<https://devfeed.tech/tags/databases.md>), [edition-us](<https://devfeed.tech/tags/edition-us.md>), [enterprises](<https://devfeed.tech/tags/enterprises.md>), [exception](<https://devfeed.tech/tags/exception.md>), [foundry](<https://devfeed.tech/tags/foundry.md>), [gpu](<https://devfeed.tech/tags/gpu.md>), [hybrid-cloud](<https://devfeed.tech/tags/hybrid-cloud.md>), [language-en](<https://devfeed.tech/tags/language-en.md>), [manufacturing](<https://devfeed.tech/tags/manufacturing.md>), [migration](<https://devfeed.tech/tags/migration.md>), [models](<https://devfeed.tech/tags/models.md>), [performance](<https://devfeed.tech/tags/performance.md>), [private-cloud](<https://devfeed.tech/tags/private-cloud.md>), [prompts](<https://devfeed.tech/tags/prompts.md>), [public-cloud](<https://devfeed.tech/tags/public-cloud.md>), [run-ai](<https://devfeed.tech/tags/run-ai.md>), [security](<https://devfeed.tech/tags/security.md>), [storytype-feature](<https://devfeed.tech/tags/storytype-feature.md>)

### AI overview

Cloud repatriation is selective: although 74% of surveyed IT decision-makers say migration accelerated, 66% say their organizations have moved or plan to move workloads out of public cloud. Organizations have moved back an average of 22% of workloads and plan another 24%; cloud-first remains the default for Protiviti clients. Security leads stated reasons (60%), followed by cost and performance (42% each) and sovereignty (40%). AI workloads are also moving: 34% have already been moved out of public cloud and 38% are planned. Placement depends on workload and operating conditions: high, predictable utilization can favor private infrastructure, while intermittent demand, extra capacity needs, and specialized capabilities can favor public cloud. The text also weighs sovereignty and control, GPU scarcity and sizing risk, model availability, and provider trust. It anticipates measured, business-value-driven decisions across public, private, sovereign, and edge environments, with governance and economics as ongoing challenges.

### Source excerpt

Enterprises are moving to the cloud faster than they have in years. Nearly three-quarters (74%) of IT decision-makers say their organizations sped up cloud migration over the past 12 months, according to Foundry's 2026 Cloud Computing Study. Yet a growing number are also bringing some workloads back in-house. About two-thirds (66%) say their organizations have moved or plan to move workloads out of public cloud. "Repatriation is one part of a wider rethink of where workloads belong," says Mukesh Ranjan, vice president at Everest Group, a research and advisory firm. AI workloads are part of the calculus. They're leaving public cloud nearly as often as storage and databases, and just 13% of IT decision-makers plan to run AI primarily there, according to Foundry's research. For CIOs, that means deciding workload by workload where each one belongs, with AI adding new factors to weigh, from GPU costs to where models run and where prompts are processed. Foundry Selective moves Foundry's numbers don't suggest a mass exodus, however. On average, organizations have moved 22% of their workloads back from public cloud and plan to move another 24% in the next 12 months. One-third of organizations have moved at least a quarter of their workloads back on-premises, down from 37% in 2024. Many organizations moving workloads out of public cloud may be pulling back only two or three, Ranjan says. They're certainly not rejecting the public cloud outright. For global consulting firm Protiviti's clients, "cloud first" remains the default, says Randy Armknecht, managing director there. "Repatriation is still the exception, driven by a thorough analysis of business value and technical capability," he says. Where AI fits in Security concerns top the reasons organizations give for moving workloads out of public cloud (60%), followed by cost and performance (42% each) and sovereignty (40%), according to the Foundry survey. "Outside manufacturing, AI is a passenger on the repatriation trend d

## You've decided to repatriate cloud workloads. Now what?

DevFeed: [You've decided to repatriate cloud workloads. Now what?](<https://devfeed.tech/articles/you-ve-decided-to-repatriate-cloud-workloads-now-what-137074.md>)

Original publisher: [Read original article](<https://www.networkworld.com/article/4247951/youve-decided-to-repatriate-cloud-workloads-now-what.html>)

Author: Zeus Kerravala

Published: 2026-11-17T10:00:00Z

Content type: article

Language: en

Sources: [Network World](<https://devfeed.tech/sources/network-world.md>)

Topics: [cloud-migration](<https://devfeed.tech/topics/cloud-migration.md>), [Cloud](<https://devfeed.tech/topics/cloud.md>), [AI Infrastructure](<https://devfeed.tech/topics/ai-infrastructure.md>), [cloud cost management](<https://devfeed.tech/topics/cloud-cost-management.md>), [digital sovereignty](<https://devfeed.tech/topics/digital-sovereignty.md>), [cloud security](<https://devfeed.tech/topics/cloud-security.md>), [control-plane](<https://devfeed.tech/topics/control-plane.md>)

Tags: [2026](<https://devfeed.tech/tags/2026.md>), [ai](<https://devfeed.tech/tags/ai.md>), [applications](<https://devfeed.tech/tags/applications.md>), [budget](<https://devfeed.tech/tags/budget.md>), [cloud](<https://devfeed.tech/tags/cloud.md>), [cloud-computing](<https://devfeed.tech/tags/cloud-computing.md>), [cloud-computing-data-center-private-cloud](<https://devfeed.tech/tags/cloud-computing-data-center-private-cloud.md>), [cloud-spending](<https://devfeed.tech/tags/cloud-spending.md>), [cloud-workloads](<https://devfeed.tech/tags/cloud-workloads.md>), [colocation](<https://devfeed.tech/tags/colocation.md>), [compliance](<https://devfeed.tech/tags/compliance.md>), [control](<https://devfeed.tech/tags/control.md>), [cost](<https://devfeed.tech/tags/cost.md>), [data-center](<https://devfeed.tech/tags/data-center.md>), [data-ownership](<https://devfeed.tech/tags/data-ownership.md>), [data-sovereignty](<https://devfeed.tech/tags/data-sovereignty.md>), [edition-us](<https://devfeed.tech/tags/edition-us.md>), [enterprise](<https://devfeed.tech/tags/enterprise.md>), [enterprise-it](<https://devfeed.tech/tags/enterprise-it.md>), [hardware](<https://devfeed.tech/tags/hardware.md>), [hybrid](<https://devfeed.tech/tags/hybrid.md>), [it](<https://devfeed.tech/tags/it.md>), [language-en](<https://devfeed.tech/tags/language-en.md>), [private-cloud](<https://devfeed.tech/tags/private-cloud.md>), [storytype-feature](<https://devfeed.tech/tags/storytype-feature.md>)

### AI overview

Foundry's 2026 survey of 527 IT decision makers reports that 66% of organizations have moved or plan to move workloads from public cloud to on-premises or colocation, while 69% expect cloud spending to rise. Respondents cite security (60%), cost control and performance/reliability (42% each), and data sovereignty/privacy (40%) as reasons. Storage, databases, and web apps are common early moves; AI is also shifting toward hybrid and private infrastructure, though bursty training may remain in cloud. Repatriation brings capacity and power constraints, network integration and latency challenges, portability and compatibility work, skills gaps, and more complex operations. The article recommends securing space and power early, designing a unified hybrid network, planning AI infrastructure separately, preserving cloud-like self-service and cost visibility, and cross-training staff.

### Source excerpt

For 15 years, enterprise IT moved in one direction: out of the data center and into the public cloud. Hardware teams were trimmed, leases weren't renewed, and "cloud first" became "cloud only." That era is ending, not because the cloud failed, but because many organizations never made a placement decision. They made a migration decision and assumed it would fit everything. AI, global macro issues, data ownership, and other trends are prompting IT leadership to rethink their cloud strategy. New data from Foundry's 2026 Cloud Computing Study, which surveyed 527 IT decision makers, shows how far the pendulum has swung. Two-thirds (66%) of respondents say their organizations have moved or plan to move applications or workloads out of the public cloud to on-premises or colocation locations. On average, respondents report that about 22% of their public cloud workloads have already been moved back and expect to move about 24% over the next 12 months. What gets lost in the "cloud is dead" headlines is that the same study shows 69% of organizations expect cloud spending to increase, and 74% default to the cloud when buying new capabilities. Repatriation isn't a retreat from the cloud. It's a rebalancing, and a hybrid estate is fundamentally harder to run than either an all-cloud or all-on-prem estate. The strategic "why?" is one important question (See our related CIO feature: Cloud repatriation on the rise but tricky questions remain). For the people who rack the gear, light the circuits and keep things running, the question is "how?" Why workloads are coming back The top driver may surprise those who remember that security was the primary objection to moving to the cloud in the first place. Sixty percent of respondents cite security concerns as a reason for moving workloads out of the public cloud, followed by budget and cost control (42%), performance and reliability (42%), and data sovereignty and privacy (40%). Compliance comes in at 31%. That's a control story more tha

## Routing Beyond Task Categories: A Cost-Aware LAB-Bench Study

DevFeed: [Routing Beyond Task Categories: A Cost-Aware LAB-Bench Study](<https://devfeed.tech/articles/routing-beyond-task-categories-a-cost-aware-lab-bench-study-137608.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2611.20732>)

Author: Ahmed Tariq (Independent researcher), Leo Guinan (Idea Nexus Ventures)

Published: 2026-11-17T05:00:00Z

Content type: news

Language: en

Sources: [arXiv: Artificial Intelligence](<https://devfeed.tech/sources/arxiv-artificial-intelligence.md>)

Topics: [AI Chat](<https://devfeed.tech/topics/ai-chat.md>), [Machine learning](<https://devfeed.tech/topics/machine-learning.md>), [cloud cost management](<https://devfeed.tech/topics/cloud-cost-management.md>)

Tags: [artificial-intelligence](<https://devfeed.tech/tags/artificial-intelligence.md>), [astra](<https://devfeed.tech/tags/astra.md>), [bench](<https://devfeed.tech/tags/bench.md>), [budget](<https://devfeed.tech/tags/budget.md>), [computer-science](<https://devfeed.tech/tags/computer-science.md>), [cost](<https://devfeed.tech/tags/cost.md>), [cs-ai](<https://devfeed.tech/tags/cs-ai.md>), [development](<https://devfeed.tech/tags/development.md>), [evaluation](<https://devfeed.tech/tags/evaluation.md>), [features](<https://devfeed.tech/tags/features.md>), [gemini](<https://devfeed.tech/tags/gemini.md>), [gpt](<https://devfeed.tech/tags/gpt.md>), [haiku](<https://devfeed.tech/tags/haiku.md>), [inference](<https://devfeed.tech/tags/inference.md>), [lab](<https://devfeed.tech/tags/lab.md>), [model](<https://devfeed.tech/tags/model.md>), [policy](<https://devfeed.tech/tags/policy.md>), [routing](<https://devfeed.tech/tags/routing.md>), [study](<https://devfeed.tech/tags/study.md>), [target](<https://devfeed.tech/tags/target.md>)

### AI overview

In this LAB-Bench study, semantic routing and answer-based deferral both met the USD 0.678-per-1,000-query inference budget, scoring 86/144 and 87/144 on 144 additional questions. Their gains over a 95% Haiku/5% GPT mixture were modest, with paired 95% intervals spanning zero. Escalating every Haiku nonanswer scored 92/144 but exceeded budget at USD 0.710; retrospective alternatives yielded expected scores of 85.63/144 and 86.14/144. In a later 40-question development check, Astra and Gemini each scored 30/40, matching the best policy in each router family; together they covered 33/40 in hindsight. Since the routers missed the 32/40 criterion, the planned 120-question evaluation did not proceed.

### Source excerpt

arXiv:2611.20732v1 Announce Type: new Abstract: We tested whether features of biology questions or of a model's first answer can improve accuracy within an inference budget. In a LAB-Bench pilot, we compared a category router with one that adds six structural features. Neither reached the primary 80% accuracy target on development. At the 50% and 60% secondary targets, both chose Haiku for all 120 held-out questions and scored 71/120. We then froze a new comparison before collecting responses on 144 additional questions. Semantic routing scored 86/144. A policy that could defer after reading Haiku's answer scored 87/144. Both stayed within the inference budget of USD 0.678 per 1,000 queries. Against a 95% Haiku and 5% GPT mixture, the accuracy gains were 0.66 percentage points for semantic routing and 1.35 points for deferral. The paired 95% intervals were [-0.28,2.25] and [-0.14,3.47] points. A simple rule that escalated Haiku nonanswers scored 92/144 at USD 0.710 per 1,000 queries, above the budget. A retrospective nonanswer rule with a calibration-selected escalation probability gave an expected 85.63/144. At the learned deferral policy's observed inference cost, post hoc interpolation gave an expected 86.14/144. In a later development check, Astra and Gemini each scored 30/40. The best tested policy in each router family matched that score. The pair covered 33/40 in hindsight. The routers missed the 32/40 criterion, so the proposed 120-question evaluation did not start.

## Announcing updates to the AWS Well-Architected Framework

DevFeed: [Announcing updates to the AWS Well-Architected Framework](<https://devfeed.tech/articles/announcing-updates-to-the-aws-well-architected-framework-136531.md>)

Original publisher: [Read original article](<https://aws.amazon.com/blogs/architecture/announcing-updates-to-the-aws-well-architected-framework-3/>)

Author: Haleh Najafzadeh

Published: 2026-11-16T20:03:26Z

Content type: release

Language: en

Sources: [AWS Architecture Blog](<https://devfeed.tech/sources/aws-architecture-blog.md>)

Topics: [Cloud Architecture](<https://devfeed.tech/topics/cloud-architecture.md>), [Amazon Web Services](<https://devfeed.tech/topics/aws.md>), [cloud cost management](<https://devfeed.tech/topics/cloud-cost-management.md>), [Amazon Bedrock](<https://devfeed.tech/topics/amazon-bedrock.md>)

Tags: [2026](<https://devfeed.tech/tags/2026.md>), [access-management](<https://devfeed.tech/tags/access-management.md>), [account](<https://devfeed.tech/tags/account.md>), [agentic](<https://devfeed.tech/tags/agentic.md>), [agentic-ai](<https://devfeed.tech/tags/agentic-ai.md>), [agents](<https://devfeed.tech/tags/agents.md>), [ai](<https://devfeed.tech/tags/ai.md>), [ai-agents](<https://devfeed.tech/tags/ai-agents.md>), [ai-assistants](<https://devfeed.tech/tags/ai-assistants.md>), [announcements](<https://devfeed.tech/tags/announcements.md>), [api](<https://devfeed.tech/tags/api.md>), [aws](<https://devfeed.tech/tags/aws.md>), [aws-well-architected](<https://devfeed.tech/tags/aws-well-architected.md>), [aws-well-architected-agent](<https://devfeed.tech/tags/aws-well-architected-agent.md>), [aws-well-architected-framework](<https://devfeed.tech/tags/aws-well-architected-framework.md>), [best-practices](<https://devfeed.tech/tags/best-practices.md>), [cloud](<https://devfeed.tech/tags/cloud.md>), [cloud-workloads](<https://devfeed.tech/tags/cloud-workloads.md>), [foundational-100](<https://devfeed.tech/tags/foundational-100.md>), [foundations](<https://devfeed.tech/tags/foundations.md>), [framework](<https://devfeed.tech/tags/framework.md>), [updates](<https://devfeed.tech/tags/updates.md>)

### AI overview

The update separates environment-wide guidance into Well-Architected Cloud Foundations and workload-specific guidance into Well-Architected Cloud Workloads, so teams assess shared guardrails once and keep workload reviews focused. Foundations has 45 best practices across seven capabilities; Workloads has 216, reduced from 307 through relocation, consolidation, and retirement of nontechnical practices, with the article stating technical coverage is retained. New guidance includes centralized event monitoring and DNS management, secure coding, cyber resilience for backup data, generative AI for operational efficiency, and resource-efficiency measurement. Operational Excellence guidance on AI assistants and agents calls for validation against baselines and named human ownership. The article also says the AWS Well-Architected Agent is generally available to analyze workloads and guide remediation.

### Source excerpt

Updates to the AWS Well-Architected Framework reorganize it into Well-Architected Cloud Foundations and Well-Architected Cloud Workloads, add new best practices, and bring the general availability of the AWS Well-Architected Agent for analyzing workloads and guiding remediation.

## Cloud overspend is an accountability problem not a tooling gap

DevFeed: [Cloud overspend is an accountability problem not a tooling gap](<https://devfeed.tech/articles/cloud-overspend-is-an-accountability-problem-not-a-tooling-gap-136138.md>)

Original publisher: [Read original article](<https://www.networkworld.com/article/4249100/cloud-overspend-is-an-accountability-problem-not-a-tooling-gap.html>)

Author: Sean Michael Kerner

Published: 2026-11-16T19:49:24Z

Content type: news

Language: en

Sources: [Network World](<https://devfeed.tech/sources/network-world.md>)

Topics: [finops](<https://devfeed.tech/topics/finops.md>), [cloud cost management](<https://devfeed.tech/topics/cloud-cost-management.md>), [Cloud](<https://devfeed.tech/topics/cloud.md>), [The Three Main Cloud Service Models](<https://devfeed.tech/topics/the-three-main-cloud-service-models.md>), [Software as a service](<https://devfeed.tech/topics/saas.md>), [Infrastructure as a Service (IaaS)](<https://devfeed.tech/topics/infrastructure-as-a-service-iaas.md>), [Platform as a Service (PaaS)](<https://devfeed.tech/topics/platform-as-a-service-paas.md>)

Tags: [17](<https://devfeed.tech/tags/17.md>), [2026](<https://devfeed.tech/tags/2026.md>), [accelerated](<https://devfeed.tech/tags/accelerated.md>), [accountability](<https://devfeed.tech/tags/accountability.md>), [adoption](<https://devfeed.tech/tags/adoption.md>), [automation](<https://devfeed.tech/tags/automation.md>), [availability](<https://devfeed.tech/tags/availability.md>), [change](<https://devfeed.tech/tags/change.md>), [cloud](<https://devfeed.tech/tags/cloud.md>), [cloud-computing](<https://devfeed.tech/tags/cloud-computing.md>), [cloud-computing-network-management-software-networking](<https://devfeed.tech/tags/cloud-computing-network-management-software-networking.md>), [cloud-operations](<https://devfeed.tech/tags/cloud-operations.md>), [control](<https://devfeed.tech/tags/control.md>), [cost](<https://devfeed.tech/tags/cost.md>), [design](<https://devfeed.tech/tags/design.md>), [doing](<https://devfeed.tech/tags/doing.md>), [edition-us](<https://devfeed.tech/tags/edition-us.md>), [flexera](<https://devfeed.tech/tags/flexera.md>), [gap](<https://devfeed.tech/tags/gap.md>), [governance](<https://devfeed.tech/tags/governance.md>), [iaas](<https://devfeed.tech/tags/iaas.md>), [infrastructure](<https://devfeed.tech/tags/infrastructure.md>), [language-en](<https://devfeed.tech/tags/language-en.md>), [network-management-software](<https://devfeed.tech/tags/network-management-software.md>), [networking](<https://devfeed.tech/tags/networking.md>), [storytype-news](<https://devfeed.tech/tags/storytype-news.md>), [tooling](<https://devfeed.tech/tags/tooling.md>)

### AI overview

Cloud overspend is framed as an accountability and operating-model problem: teams remain responsible for availability, cost, and risk while having less control over cloud mechanisms. Sherif cites Flexera's estimate that 29% of IaaS and PaaS spend is wasted and public cloud budgets run 17% over on average. He argues that more tools, training, self-service, or automation do not resolve responsibility without leverage. His method is to classify work as consuming SaaS, hosting workloads, or building custom services; prioritize the area with the greatest risk or volume; define the work areas and suitable governance or delivery patterns; then map current teams to identify gaps. SaaS-heavy organizations need vendor management and integration skills, while development-heavy ones need CI/CD pipelines and developer support. Cost, reliability, and security should be considered at design time. The supplied text presents this as Sherif's approach, not a universal prescription.

### Source excerpt

After years of using the clouds, many organizations are coming to realize they didn't have a proper plan in place from the get-go. Cloud adoption has often been driven by urgency, not design, according to Nabeel Sherif, principal advisory director at Infotech. It accelerated over the past 15 years, especially during the pandemic. Most organizations moved workloads and adopted SaaS, but the operating models behind that work did not change at the same pace. Infrastructure and operations teams, network teams included, now run on-premises and cloud side by side, often without a unifying model. They remain accountable for availability, cost, and risk but with far less direct control than in the past. The cost shows up in the numbers. Flexera's 2026 State of the Cloud Report, which Sherif cited, estimates that 29% of IaaS and PaaS spend is wasted. Organizations also run 17% over their public cloud budgets on average. During a session at the Infotech LIVE 2026 conference in Toronto, Sherif outlined a method for redesigning cloud operations around the work that needs doing, then assigning that work to teams. "Cloud doesn't eliminate operations work; it just changes its nature," Sherif said. Why more tools and training miss the problem The common reaction that many organizations have to cloud problems is to buy more tools or add training. Those actions don't however always solve the underlying issues. In Sherif's view, the data points to governance, organizational structure, visibility, and accountability as the right path forward. Accountability is the sticking point because ownership and control no longer line up in the cloud. "In traditional operations, you assume ownership of the stack. Cloud breaks that assumption," Sherif said. "You still own the outcomes. You don't necessarily own all the underlying mechanisms." He added that neither self-service nor automation fixes the issues. "Without redesign, you get responsibility without leverage," he said. "So, governance, cos

## Gartner Analyst: "Your Next Database User Won't Be a Person"

DevFeed: [Gartner Analyst: "Your Next Database User Won't Be a Person"](<https://devfeed.tech/articles/gartner-analyst-your-next-database-user-won-t-be-a-person-137751.md>)

Original publisher: [Read original article](<https://flashdba.com/2026/11/16/gartner-analyst-your-next-database-user-wont-be-a-person/>)

Author: flashdba

Published: 2026-11-16T17:27:25Z

Content type: opinion

Language: en

Sources: [Flashdba](<https://devfeed.tech/sources/flashdba.md>)

Topics: [Database](<https://devfeed.tech/topics/database.md>), [cloud cost management](<https://devfeed.tech/topics/cloud-cost-management.md>), [database-scalability](<https://devfeed.tech/topics/database-scalability.md>), [read-replica](<https://devfeed.tech/topics/read-replica.md>), [database monitoring](<https://devfeed.tech/topics/database-monitoring.md>), [connection pool](<https://devfeed.tech/topics/connection-pool.md>)

Tags: [2026](<https://devfeed.tech/tags/2026.md>), [about](<https://devfeed.tech/tags/about.md>), [account](<https://devfeed.tech/tags/account.md>), [after](<https://devfeed.tech/tags/after.md>), [age-of-ai](<https://devfeed.tech/tags/age-of-ai.md>), [agent](<https://devfeed.tech/tags/agent.md>), [agentic](<https://devfeed.tech/tags/agentic.md>), [agents](<https://devfeed.tech/tags/agents.md>), [ai](<https://devfeed.tech/tags/ai.md>), [ai-agents](<https://devfeed.tech/tags/ai-agents.md>), [alarm](<https://devfeed.tech/tags/alarm.md>), [budget](<https://devfeed.tech/tags/budget.md>), [cloud](<https://devfeed.tech/tags/cloud.md>), [database](<https://devfeed.tech/tags/database.md>), [databases](<https://devfeed.tech/tags/databases.md>), [databases-and-agentic-ai](<https://devfeed.tech/tags/databases-and-agentic-ai.md>), [first](<https://devfeed.tech/tags/first.md>), [gartner](<https://devfeed.tech/tags/gartner.md>), [governance](<https://devfeed.tech/tags/governance.md>), [inferencing](<https://devfeed.tech/tags/inferencing.md>), [next](<https://devfeed.tech/tags/next.md>), [operational](<https://devfeed.tech/tags/operational.md>), [performance](<https://devfeed.tech/tags/performance.md>), [user](<https://devfeed.tech/tags/user.md>)

### AI overview

The excerpt argues that gradual AI agent adoption is more likely to expose budget and governance problems before cloud operational databases fail. Much agent activity is read-heavy, and cloud platforms can absorb added load with replicas, serverless compute, or higher provisioned limits; this can keep queries and latency healthy while shifting the strain into infrastructure costs, often charged to teams that did not deploy the agents. Database failure can still occur beyond elasticity limits, but the article says organizations may pay for the approach well before that point. Governance is a separate, pre-existing gap: an agent using a service account may be identifiable only as that account, leaving records without attribution to the agent, its initiator, authority, or purpose. Higher volume multiplies these gaps without necessarily triggering an alert. The excerpt also notes that database choices are moving from central DBA teams to product and engineering leads as DBA headcount declines, while system owners may still receive the bill and audit findings.

### Source excerpt

A Gartner analyst asked what AI agents would break first. On cloud operational databases the budget breaks before the database - and governance already has.

## Set budgets for projects and products in Spend Management

DevFeed: [Set budgets for projects and products in Spend Management](<https://devfeed.tech/articles/set-budgets-for-projects-and-products-in-spend-management-136324.md>)

Original publisher: [Read original article](<https://vercel.com/changelog/set-budgets-for-projects-and-products-in-spend-management>)

Author: Lakshmi Subbramanian

Published: 2026-11-16T00:00:00Z

Content type: release

Language: en

Sources: [Vercel News](<https://devfeed.tech/sources/vercel-news.md>)

Topics: [cloud cost management](<https://devfeed.tech/topics/cloud-cost-management.md>)

Tags: [billing](<https://devfeed.tech/tags/billing.md>), [budget](<https://devfeed.tech/tags/budget.md>), [cli](<https://devfeed.tech/tags/cli.md>), [deployments](<https://devfeed.tech/tags/deployments.md>), [documentation](<https://devfeed.tech/tags/documentation.md>), [management](<https://devfeed.tech/tags/management.md>), [notifications](<https://devfeed.tech/tags/notifications.md>), [other](<https://devfeed.tech/tags/other.md>), [production](<https://devfeed.tech/tags/production.md>), [products](<https://devfeed.tech/tags/products.md>), [project](<https://devfeed.tech/tags/project.md>), [projects](<https://devfeed.tech/tags/projects.md>), [spend-management](<https://devfeed.tech/tags/spend-management.md>), [team](<https://devfeed.tech/tags/team.md>), [vercel](<https://devfeed.tech/tags/vercel.md>)

### AI overview

Vercel Spend Management now lets teams set budgets for individual projects and products alongside team budgets. Project budgets can optionally pause that project's production deployments when the budget is reached; product budgets track spend on a selected product across the team. Automatic deployment pausing is available only for team and project budgets. The Spend Management page shows current billing-cycle spend and reached thresholds, with notifications at 50%, 75%, and 100%. Budgets can also be listed or set with the Vercel CLI. The feature is included at no additional cost for Pro teams and Enterprise teams on Flexible Commitment plans; Organizations private beta customers can set an organization-wide budget across teams.

### Source excerpt

Spend Management now supports budgets for individual projects and products, alongside existing team budgets. Project budgets: Track spend for one project and optionally pause its production deployments when the budget is reached, without pausing other projects. Product budgets: Track spend on one product, such as Fast Data Transfer, across your team. Only team and project budgets support automatic pausing of production deployments. Open the new Spend Management page to manage budgets, view spend for the current billing cycle, and see which thresholds have been reached. Vercel sends notifications at 50%, 75%, and 100% of each budget. You can also manage budgets with the Vercel CLI: Spend Management is included at no additional cost for Pro teams and Enterprise teams on Flexible Commitment plans. Customers in the Organizations private beta can also set a budget across all teams in their organization. Learn more in the Spend Management documentation. Read more

## AI Coding Cost Attribution: From Tokens to Features

DevFeed: [AI Coding Cost Attribution: From Tokens to Features](<https://devfeed.tech/articles/ai-coding-cost-attribution-from-tokens-to-features-136340.md>)

Original publisher: [Read original article](<https://www.harness.io/blog/ai-coding-cost-attribution-tokens-to-features>)

Author: Kelsey Rosen

Published: 2026-11-16T00:00:00Z

Content type: article

Language: en

Sources: [Harness](<https://devfeed.tech/sources/harness-blog.md>)

Topics: [LLM observability](<https://devfeed.tech/topics/llm-observability.md>), [coding assistant](<https://devfeed.tech/topics/coding-assistant.md>), [agentic-coding](<https://devfeed.tech/topics/agentic-coding.md>), [cloud cost management](<https://devfeed.tech/topics/cloud-cost-management.md>), [GitHub Copilot](<https://devfeed.tech/topics/github-copilot.md>), [Claude Code](<https://devfeed.tech/topics/claude-code.md>), [cursor](<https://devfeed.tech/topics/cursor.md>)

Tags: [agent](<https://devfeed.tech/tags/agent.md>), [agent-harness](<https://devfeed.tech/tags/agent-harness.md>), [aggregates](<https://devfeed.tech/tags/aggregates.md>), [ai](<https://devfeed.tech/tags/ai.md>), [ai-assisted](<https://devfeed.tech/tags/ai-assisted.md>), [ai-coding](<https://devfeed.tech/tags/ai-coding.md>), [ai-coding-tools](<https://devfeed.tech/tags/ai-coding-tools.md>), [ai-data](<https://devfeed.tech/tags/ai-data.md>), [ai-engineering](<https://devfeed.tech/tags/ai-engineering.md>), [attribution](<https://devfeed.tech/tags/attribution.md>), [blog](<https://devfeed.tech/tags/blog.md>), [coding](<https://devfeed.tech/tags/coding.md>), [commits](<https://devfeed.tech/tags/commits.md>), [connect](<https://devfeed.tech/tags/connect.md>), [cost](<https://devfeed.tech/tags/cost.md>), [features](<https://devfeed.tech/tags/features.md>), [how-to](<https://devfeed.tech/tags/how-to.md>), [pull-requests](<https://devfeed.tech/tags/pull-requests.md>), [tickets](<https://devfeed.tech/tags/tickets.md>), [tokens](<https://devfeed.tech/tags/tokens.md>)

### AI overview

The webinar recap recommends rolling AI coding spend up to features as a practical starting point: provider token counts show consumption, not whether generated code reached production or delivered value, and finer span-level attribution may add diminishing returns. Trimble pairs cost with velocity and quality, uses counter-metrics and team baselines to limit metric gaming, and separates experimentation and duplicate tool use from production spend. Messy links among PRs, tickets, commits, and service accounts make attribution approximate without engineering discipline; Trimble layers AI data onto Harness Engineering Efficiency. The speakers also describe role-based budgets with a fast increase request path, historical baselines for rough cost estimates, and a platform-plus-custom-context approach. They say reliable token cost prediction per feature remains unsolved.

### Source excerpt

Martin Reynolds and Trimble's Barani Mohandoss explain how to connect AI coding spend to commits, pull requests, tickets, and shipped features. | Blog

## Cost-optimizing AssemblyAI speech-to-text inference on AWS with Amazon EC2 Spot

DevFeed: [Cost-optimizing AssemblyAI speech-to-text inference on AWS with Amazon EC2 Spot](<https://devfeed.tech/articles/cost-optimizing-assemblyai-speech-to-text-inference-on-aws-with-amazon-ec2-spot-132652.md>)

Original publisher: [Read original article](<https://aws.amazon.com/blogs/machine-learning/cost-optimizing-assemblyai-speech-to-text-inference-on-aws-with-amazon-ec2-spot/>)

Author: Belair Albahagwi

Published: 2026-11-13T19:35:35Z

Content type: news

Language: en

Sources: [Artificial Intelligence](<https://devfeed.tech/sources/artificial-intelligence.md>)

Topics: [asr](<https://devfeed.tech/topics/asr.md>), [AI Infrastructure](<https://devfeed.tech/topics/ai-infrastructure.md>), [Amazon EC2](<https://devfeed.tech/topics/amazon-ec2.md>), [cloud cost management](<https://devfeed.tech/topics/cloud-cost-management.md>), [Amazon Web Services](<https://devfeed.tech/topics/aws.md>), [Speech Processing](<https://devfeed.tech/topics/speech-processing.md>), [AI Architecture](<https://devfeed.tech/topics/ai-architecture.md>)

Tags: [ai-inference](<https://devfeed.tech/tags/ai-inference.md>), [amazon-ec2](<https://devfeed.tech/tags/amazon-ec2.md>), [amazon-sqs](<https://devfeed.tech/tags/amazon-sqs.md>), [architecture](<https://devfeed.tech/tags/architecture.md>), [assemblyai](<https://devfeed.tech/tags/assemblyai.md>), [aws](<https://devfeed.tech/tags/aws.md>), [cost](<https://devfeed.tech/tags/cost.md>), [cost-savings](<https://devfeed.tech/tags/cost-savings.md>), [customer-solutions](<https://devfeed.tech/tags/customer-solutions.md>), [ec2](<https://devfeed.tech/tags/ec2.md>), [ecs](<https://devfeed.tech/tags/ecs.md>), [gpu](<https://devfeed.tech/tags/gpu.md>), [inference](<https://devfeed.tech/tags/inference.md>), [optimized](<https://devfeed.tech/tags/optimized.md>), [regions](<https://devfeed.tech/tags/regions.md>), [retry](<https://devfeed.tech/tags/retry.md>), [scale](<https://devfeed.tech/tags/scale.md>), [speech](<https://devfeed.tech/tags/speech.md>), [speech-to-text](<https://devfeed.tech/tags/speech-to-text.md>), [spot](<https://devfeed.tech/tags/spot.md>), [sqs](<https://devfeed.tech/tags/sqs.md>)

### AI overview

AssemblyAI reports 70% lower inference compute costs than On-Demand pricing using GPU Spot Instances, while maintaining SLA commitments. Its Amazon ECS clusters combine an always-on On-Demand baseline with diversified Spot capacity across g4dn, g5, and g6 GPU families and Availability Zones. Amazon SQS decouples job submission from execution: tasks finish work within Spot's two-minute interruption notice or return unfinished jobs to the queue for another worker to retry. Custom images with model artifacts and containers take about two minutes to create, reducing recovery time. Independent deployments in each AWS Region keep audio within required geographic boundaries. The article says compute demand grew nearly fourfold year over year; interruptions may add processing time, and exhausted retries can be routed to a dead-letter queue for investigation.

### Source excerpt

Learn how AssemblyAI runs cost-optimized speech-to-text inference at scale on AWS, using Amazon ECS, Amazon EC2 Spot GPU instances, and an Amazon SQS-based retry architecture to achieve 70% cost savings while maintaining SLA commitments across European Regions.

## Cloud reports measure the wrong thing

DevFeed: [Cloud reports measure the wrong thing](<https://devfeed.tech/articles/cloud-reports-measure-the-wrong-thing-133214.md>)

Original publisher: [Read original article](<https://www.infoworld.com/article/4248511/cloud-reports-measure-the-wrong-thing.html>)

Author: David Linthicum

Published: 2026-11-13T16:21:12Z

Content type: opinion

Language: en

Sources: [InfoWorld](<https://devfeed.tech/sources/infoworld.md>)

Topics: [cloud cost management](<https://devfeed.tech/topics/cloud-cost-management.md>), [finops](<https://devfeed.tech/topics/finops.md>), [Cloud Architecture](<https://devfeed.tech/topics/cloud-architecture.md>), [cost-optimization](<https://devfeed.tech/topics/cost-optimization.md>), [ai-governance](<https://devfeed.tech/topics/ai-governance.md>)

Tags: [2026](<https://devfeed.tech/tags/2026.md>), [ai](<https://devfeed.tech/tags/ai.md>), [ai-usage](<https://devfeed.tech/tags/ai-usage.md>), [analysts](<https://devfeed.tech/tags/analysts.md>), [cloud](<https://devfeed.tech/tags/cloud.md>), [cloud-architecture](<https://devfeed.tech/tags/cloud-architecture.md>), [cloud-architecture-cloud-computing](<https://devfeed.tech/tags/cloud-architecture-cloud-computing.md>), [cloud-computing](<https://devfeed.tech/tags/cloud-computing.md>), [cloud-costs](<https://devfeed.tech/tags/cloud-costs.md>), [enterprise](<https://devfeed.tech/tags/enterprise.md>), [finops](<https://devfeed.tech/tags/finops.md>), [flexera](<https://devfeed.tech/tags/flexera.md>), [generative-ai](<https://devfeed.tech/tags/generative-ai.md>), [governance](<https://devfeed.tech/tags/governance.md>), [hybrid-cloud](<https://devfeed.tech/tags/hybrid-cloud.md>), [iaas](<https://devfeed.tech/tags/iaas.md>), [language-en](<https://devfeed.tech/tags/language-en.md>), [metric](<https://devfeed.tech/tags/metric.md>), [organizations](<https://devfeed.tech/tags/organizations.md>), [public-cloud](<https://devfeed.tech/tags/public-cloud.md>), [report](<https://devfeed.tech/tags/report.md>), [reports](<https://devfeed.tech/tags/reports.md>), [storytype-opinion](<https://devfeed.tech/tags/storytype-opinion.md>), [survey](<https://devfeed.tech/tags/survey.md>), [team](<https://devfeed.tech/tags/team.md>)

### AI overview

The article argues that cloud survey metrics can make management activity look like maturity while obscuring architectural debt and weak business cases. It cites Flexera's survey of 753 cloud decision-makers: 64% measure value delivered to business units, 81% still prioritize cost efficiency and savings, 29% estimate wasted IaaS/PaaS spend, and 58% report using generative AI public cloud services. The author cautions that hybrid cloud may reflect acquisitions or inherited infrastructure rather than deliberate strategy, and that FinOps, dashboards, policies, and discounts can expose or reduce costs but cannot repair poor system design or uneconomic workload placement. It also questions whether reported "value" and AI use demonstrate measurable outcomes, noting that usage may include experiments or pilots. Because respondents are cloud-involved and the report is vendor-funded, the author recommends treating its findings as inputs rather than market truth.

### Source excerpt

Flexera's 2026 State of the Cloud Report agrees with many other cloud reports these days: Cloud is more strategic, cloud is more complex, AI makes everything harder, and better governance is the answer. This report, based on a survey of 753 cloud decision-makers and users, argues that enterprise cloud has entered a "value era" in which organizations are measuring success less by cost savings alone and more by business outcomes. The headline findings are familiar. Flexera says 64% of organizations now use "value delivered to business units" as a metric for cloud progress, up 12 percentage points year over year. Use of unit economics rose from 40% to 49%. Generative AI usage as a public cloud service rose to 58%, making it one of the most widely used public cloud services in the survey. Hybrid cloud remains dominant at 73%. Estimated wasted IaaS and PaaS cloud spend ticked up to 29%, reversing a five-year decline. Meanwhile, 71% of respondents say their company has a cloud center of excellence or a similar team, and 63% say they have a finops team. These reports provide a decent temperature check of what cloud teams are thinking about. The problem is what happens next. A survey of cloud-involved people becomes a story about the market. A set of self-reported answers becomes evidence of maturity. A vendor-funded report becomes a road map. Then CIOs, analysts, vendors, and conference speakers all repeat the same narrative until it sounds like reality. It is not always reality. The findings are not the problem I agree with Flexera that cloud complexity is increasing. I agree that AI is making cloud costs harder to predict. I agree that many enterprises are adding finops teams, governance groups, and centralized cloud functions. The issue is interpretation. The report frames much of this as cloud maturing into a more value-driven discipline. That is partly true, but it is also the optimistic version of the story. The less flattering version is that enterprises are still p

## Introducing Spend Management: Spending limits and alerts for Cloudflare

DevFeed: [Introducing Spend Management: Spending limits and alerts for Cloudflare](<https://devfeed.tech/articles/introducing-spend-management-spending-limits-and-alerts-for-cloudflare-133524.md>)

Original publisher: [Read original article](<https://blog.cloudflare.com/spend-management/>)

Author: André Cruz

Published: 2026-11-13T14:00:00Z

Content type: release

Language: en

Sources: [Cloudflare Blog](<https://devfeed.tech/sources/cloudflare-blog.md>)

Topics: [cloud cost management](<https://devfeed.tech/topics/cloud-cost-management.md>), [Cloudflare](<https://devfeed.tech/topics/cloudflare.md>), [Cloudflare Workers](<https://devfeed.tech/topics/cloudflare-workers.md>), [Developer Platform](<https://devfeed.tech/topics/developer-platform.md>)

Tags: [agentic](<https://devfeed.tech/tags/agentic.md>), [ai](<https://devfeed.tech/tags/ai.md>), [ai-agents](<https://devfeed.tech/tags/ai-agents.md>), [ai-gateway](<https://devfeed.tech/tags/ai-gateway.md>), [ai-inference](<https://devfeed.tech/tags/ai-inference.md>), [ai-requests](<https://devfeed.tech/tags/ai-requests.md>), [alerts](<https://devfeed.tech/tags/alerts.md>), [analytics](<https://devfeed.tech/tags/analytics.md>), [api](<https://devfeed.tech/tags/api.md>), [apis](<https://devfeed.tech/tags/apis.md>), [architecture](<https://devfeed.tech/tags/architecture.md>), [asynchronous](<https://devfeed.tech/tags/asynchronous.md>), [billing](<https://devfeed.tech/tags/billing.md>), [cloudflare](<https://devfeed.tech/tags/cloudflare.md>), [developer-platform](<https://devfeed.tech/tags/developer-platform.md>), [developers](<https://devfeed.tech/tags/developers.md>), [durable-objects](<https://devfeed.tech/tags/durable-objects.md>), [limits](<https://devfeed.tech/tags/limits.md>), [pipeline](<https://devfeed.tech/tags/pipeline.md>), [runtime](<https://devfeed.tech/tags/runtime.md>), [spend-management](<https://devfeed.tech/tags/spend-management.md>), [spending](<https://devfeed.tech/tags/spending.md>), [workers](<https://devfeed.tech/tags/workers.md>)

### AI overview

Cloudflare Spend Management Beta lets any pay-as-you-go customer set spending limits and alert thresholds for Workers and Durable Objects. When a cap is reached, a blocking policy pauses new cost-incurring invocations; static assets in Workers continue to be served. Customers can configure controls through the dashboard, cf CLI, or REST API, with estimated cycle costs shown in the dashboard in under 10 minutes. Billing cycles follow the paid subscription period. To resume service after a cap, raise the limit above accumulated spend, disable or delete the policy, or wait for the next cycle. Enforcement is asynchronous: aggregated usage is rated outside the request path, then caps propagate globally and runtimes reject capped operations using local state. The article says other products and shorter enforcement delays are planned; these are not yet described as available.

### Source excerpt

Spend Management adds opt-in spending caps for Cloudflare Workers and Durable Objects, and more products soon. Here is how we built the metering pipeline that turns usage into runtime decisions.

## Google Cloud takes aim at the cost of keeping idle AI agents running on Kubernetes

DevFeed: [Google Cloud takes aim at the cost of keeping idle AI agents running on Kubernetes](<https://devfeed.tech/articles/google-cloud-takes-aim-at-the-cost-of-keeping-idle-ai-agents-running-on-kubernetes-131307.md>)

Original publisher: [Read original article](<https://siliconangle.com/2026/11/13/distributed-training-inference-google-s-agent-projects-kubeconna/>)

Author: Ryan Stevens

Published: 2026-11-13T07:42:29Z

Content type: news

Language: en

Sources: [SiliconANGLE](<https://devfeed.tech/sources/siliconangle.md>)

Topics: [AI Infrastructure](<https://devfeed.tech/topics/ai-infrastructure.md>), [Google Cloud Platform (GCP)](<https://devfeed.tech/topics/google-cloud.md>), [cloud cost management](<https://devfeed.tech/topics/cloud-cost-management.md>), [Gke](<https://devfeed.tech/topics/gke.md>), [Cloud](<https://devfeed.tech/topics/cloud.md>), [Local AI](<https://devfeed.tech/topics/local-ai.md>)

Tags: [agent-substrate](<https://devfeed.tech/tags/agent-substrate.md>), [agentic](<https://devfeed.tech/tags/agentic.md>), [agentic-ai](<https://devfeed.tech/tags/agentic-ai.md>), [agentic-workloads](<https://devfeed.tech/tags/agentic-workloads.md>), [agents](<https://devfeed.tech/tags/agents.md>), [ai](<https://devfeed.tech/tags/ai.md>), [ai-agents](<https://devfeed.tech/tags/ai-agents.md>), [ai-cost](<https://devfeed.tech/tags/ai-cost.md>), [ai-inference](<https://devfeed.tech/tags/ai-inference.md>), [ai-infrastructure](<https://devfeed.tech/tags/ai-infrastructure.md>), [allan-naim](<https://devfeed.tech/tags/allan-naim.md>), [cloud](<https://devfeed.tech/tags/cloud.md>), [cloud-native](<https://devfeed.tech/tags/cloud-native.md>), [cloud-native-computing-foundation](<https://devfeed.tech/tags/cloud-native-computing-foundation.md>), [cloud-security](<https://devfeed.tech/tags/cloud-security.md>), [cloudnativecon](<https://devfeed.tech/tags/cloudnativecon.md>), [cncf](<https://devfeed.tech/tags/cncf.md>), [compute](<https://devfeed.tech/tags/compute.md>), [container](<https://devfeed.tech/tags/container.md>), [cost](<https://devfeed.tech/tags/cost.md>), [costs](<https://devfeed.tech/tags/costs.md>), [cube-event-coverage](<https://devfeed.tech/tags/cube-event-coverage.md>), [developer-platforms](<https://devfeed.tech/tags/developer-platforms.md>), [distributed](<https://devfeed.tech/tags/distributed.md>), [distributed-training](<https://devfeed.tech/tags/distributed-training.md>), [enterprise-ai](<https://devfeed.tech/tags/enterprise-ai.md>), [gari-singh](<https://devfeed.tech/tags/gari-singh.md>), [google](<https://devfeed.tech/tags/google.md>), [google-cloud](<https://devfeed.tech/tags/google-cloud.md>), [gpu-infrastructure](<https://devfeed.tech/tags/gpu-infrastructure.md>), [gvisor](<https://devfeed.tech/tags/gvisor.md>), [inference](<https://devfeed.tech/tags/inference.md>), [kubecon](<https://devfeed.tech/tags/kubecon.md>), [kubecon-plus-cloudnativecon-na-2026](<https://devfeed.tech/tags/kubecon-plus-cloudnativecon-na-2026.md>), [kubeconna](<https://devfeed.tech/tags/kubeconna.md>), [kuberay](<https://devfeed.tech/tags/kuberay.md>), [kubernetes](<https://devfeed.tech/tags/kubernetes.md>), [nasdaq-goog](<https://devfeed.tech/tags/nasdaq-goog.md>), [nasdaq-googl](<https://devfeed.tech/tags/nasdaq-googl.md>), [news](<https://devfeed.tech/tags/news.md>), [observability](<https://devfeed.tech/tags/observability.md>), [open-source](<https://devfeed.tech/tags/open-source.md>), [openrl](<https://devfeed.tech/tags/openrl.md>), [paul-nashawaty](<https://devfeed.tech/tags/paul-nashawaty.md>), [platform-engineering](<https://devfeed.tech/tags/platform-engineering.md>), [rebecca-knight](<https://devfeed.tech/tags/rebecca-knight.md>), [red-hat](<https://devfeed.tech/tags/red-hat.md>), [thecube](<https://devfeed.tech/tags/thecube.md>), [waiting](<https://devfeed.tech/tags/waiting.md>)

### AI overview

Google Cloud donated three projects to the Cloud Native Computing Foundation to address cost, security and AI workload needs on Kubernetes. Its Agent Substrate runtime hosts agents in isolated environments, snapshots their memory and local files when suspended, frees compute for other work, then restores them in an available sandbox. Google reports up to a 10x cost reduction versus running agents on dedicated virtual machines and containers. gVisor, open-sourced in 2018, sits above the host kernel to block or audit agents' system calls. KubeRay runs Anyscale's Ray platform natively on Kubernetes for distributed training and inference; OpenRL aims to bring reinforcement learning to Kubernetes platform engineering for continuous agent improvement. The excerpt says agents may wait idle 90%-95% of the time, but provides no independent validation or detail on the reported savings.

### Source excerpt

Kubernetes has become the go-to platform for distributed training and inference, but what happens to the compute bill when it runs AI agents that spend most of their time waiting? When an agent's container or virtual machine stays provisioned, customers pay for the wait as well as the work. To help customers control those costs [...] The post Google Cloud takes aim at the cost of keeping idle AI agents running on Kubernetes appeared first on SiliconANGLE.

## The Hidden Cost of Adding a CDN to Google Cloud Run

DevFeed: [The Hidden Cost of Adding a CDN to Google Cloud Run](<https://devfeed.tech/articles/the-hidden-cost-of-adding-a-cdn-to-google-cloud-run-131621.md>)

Original publisher: [Read original article](<https://hackernoon.com/the-hidden-cost-of-adding-a-cdn-to-google-cloud-run?source=rss>)

Author: Nandan Kumar

Published: 2026-11-13T07:25:15Z

Content type: article

Language: en

Sources: [HackerNoon](<https://devfeed.tech/sources/hackernoon.md>)

Topics: [cloud cost management](<https://devfeed.tech/topics/cloud-cost-management.md>), [Google Cloud Platform (GCP)](<https://devfeed.tech/topics/google-cloud.md>), [Cloud Run](<https://devfeed.tech/topics/cloud-run.md>), [Serverless](<https://devfeed.tech/topics/serverless.md>), [Cloud Architecture](<https://devfeed.tech/topics/cloud-architecture.md>)

Tags: [api](<https://devfeed.tech/tags/api.md>), [aws](<https://devfeed.tech/tags/aws.md>), [azure](<https://devfeed.tech/tags/azure.md>), [backend-development](<https://devfeed.tech/tags/backend-development.md>), [bill](<https://devfeed.tech/tags/bill.md>), [billing](<https://devfeed.tech/tags/billing.md>), [caching](<https://devfeed.tech/tags/caching.md>), [cdn](<https://devfeed.tech/tags/cdn.md>), [certificate](<https://devfeed.tech/tags/certificate.md>), [cloud](<https://devfeed.tech/tags/cloud.md>), [cloud-computing](<https://devfeed.tech/tags/cloud-computing.md>), [cloud-costs](<https://devfeed.tech/tags/cloud-costs.md>), [cloud-infrastructure](<https://devfeed.tech/tags/cloud-infrastructure.md>), [cloud-run](<https://devfeed.tech/tags/cloud-run.md>), [cost](<https://devfeed.tech/tags/cost.md>), [costs](<https://devfeed.tech/tags/costs.md>), [estimator](<https://devfeed.tech/tags/estimator.md>), [fixed](<https://devfeed.tech/tags/fixed.md>), [google-cloud](<https://devfeed.tech/tags/google-cloud.md>), [google-cloud-pricing](<https://devfeed.tech/tags/google-cloud-pricing.md>), [google-cloud-run](<https://devfeed.tech/tags/google-cloud-run.md>), [latency](<https://devfeed.tech/tags/latency.md>), [open-source](<https://devfeed.tech/tags/open-source.md>), [performance](<https://devfeed.tech/tags/performance.md>), [website](<https://devfeed.tech/tags/website.md>)

### AI overview

Adding Google Cloud CDN to a low-traffic site on request-billed Cloud Run raised the reported monthly bill from about $2 to an estimated $30. The CDN setup requires an external HTTP(S) load balancer, whose fixed monthly cost applies regardless of request volume; the author says this fixed charge drove the jump. For this site, the CDN addressed only modest latency, so the author would have weighed that benefit against the cost before deploying. The resulting open-source Cloud Cost Estimator compares configurations across Google Cloud, AWS and Azure, but uses static prices from public sources rather than live APIs, so estimates may drift and are only general guidance.

### Source excerpt

Adding Google Cloud CDN increased a website's cloud bill by 15x. Here's why it happened and how it inspired an open-source cost estimator.

## BOA Constrictor: Squeezing Performance out of GPUs in the Cloud via Budget-Optimal Allocation

DevFeed: [BOA Constrictor: Squeezing Performance out of GPUs in the Cloud via Budget-Optimal Allocation](<https://devfeed.tech/articles/boa-constrictor-squeezing-performance-out-of-gpus-in-the-cloud-via-budget-optimal-allocation-131508.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2602.01404>)

Author: Zhouzi Li, Cindy Zhu, Arpan Mukhopadhyay, Mor Harchol-Balter, Benjamin Berg

Published: 2026-11-13T05:00:00Z

Content type: article

Language: en

Sources: [arXiv: Distributed Computing](<https://devfeed.tech/sources/arxiv-distributed-computing.md>)

Topics: [scheduling](<https://devfeed.tech/topics/scheduling.md>), [distributed-training](<https://devfeed.tech/topics/distributed-training.md>), [cloud cost management](<https://devfeed.tech/topics/cloud-cost-management.md>)

Tags: [1-6](<https://devfeed.tech/tags/1-6.md>), [2](<https://devfeed.tech/tags/2.md>), [budget](<https://devfeed.tech/tags/budget.md>), [cloud](<https://devfeed.tech/tags/cloud.md>), [cs-dc](<https://devfeed.tech/tags/cs-dc.md>), [gpus](<https://devfeed.tech/tags/gpus.md>), [machine-learning](<https://devfeed.tech/tags/machine-learning.md>), [performance](<https://devfeed.tech/tags/performance.md>), [scheduler](<https://devfeed.tech/tags/scheduler.md>), [scheduling](<https://devfeed.tech/tags/scheduling.md>), [simulations](<https://devfeed.tech/tags/simulations.md>)

### AI overview

BOA Constrictor schedules cloud GPUs under a fixed budget to minimize average job completion time. Its Budget-Optimal Allocation policy jointly decides how many GPUs to rent over time and how to distribute them among active training jobs, balancing faster completion against GPU-hour costs and diminishing gains from adding GPUs. The abstract reports average JCT reductions of 1.6x in small-scale implementation experiments and 2x in large-scale simulations, compared with state-of-the-art heuristic schedulers, at a given budget.

### Source excerpt

arXiv:2602.01404v3 Announce Type: replace Abstract: The past decade has seen a dramatic increase in demand for GPUs to train Machine Learning (ML) models. Because it is prohibitively expensive for most organizations to build and maintain a large GPU cluster, organizations instead choose to rent GPUs from cloud providers. The customer is responsible for devising a policy for (i) deciding how many GPUs to rent at every moment in time to process a stream of ML training jobs and (ii) allocating the rented GPUs among the currently active jobs in the system. Because ML training jobs can be parallelized across different numbers of GPUs, the customer generally has many options for how many GPUs to use for each job. Allocating more GPUs to a single training job will cause the job to complete more quickly. However, the customer pays for each GPU-hour they use, and a training job receives a diminishing marginal benefit from running on additional GPUs. Hence, allocating too many GPUs to a single training job can dramatically increase the overall cost that the customer pays to the cloud provider. This gives rise to a cost-performance tradeoff that customers must balance when running training jobs in the cloud. To balance the cost-performance tradeoff, we develop BOA Constrictor, a new scheduler for ML training jobs which uses a Budget-Optimal Allocation (BOA) policy to squeeze the highest level of performance out of a cloud-deployed GPU cluster given a fixed budget constraint. We explicitly formulate the problem as a budget-constrained scheduling problem and derive the BOA policy which minimizes the average job completion time (JCT) of a stream of arriving jobs subject to the user's budget. For a given budget level, we demonstrate that BOA Constrictor can reduce average JCT by 1.6 times in small-scale implementation experiments and by 2 times in detailed, large-scale simulations compared to state-of-the-art heuristic based schedulers.

## Optimizing cost per task and proving agent ROI on Amazon Bedrock

DevFeed: [Optimizing cost per task and proving agent ROI on Amazon Bedrock](<https://devfeed.tech/articles/optimizing-cost-per-task-and-proving-agent-roi-on-amazon-bedrock-130771.md>)

Original publisher: [Read original article](<https://aws.amazon.com/blogs/machine-learning/optimizing-cost-per-task-and-proving-agent-roi-on-amazon-bedrock/>)

Author: Majd Arbash

Published: 2026-11-12T18:43:19Z

Content type: article

Language: en

Sources: [Artificial Intelligence](<https://devfeed.tech/sources/artificial-intelligence.md>)

Topics: [LLM optimization](<https://devfeed.tech/topics/llm-optimization.md>), [cloud cost management](<https://devfeed.tech/topics/cloud-cost-management.md>), [Multi-step AI processes](<https://devfeed.tech/topics/multi-step-ai-processes.md>), [LLM observability](<https://devfeed.tech/topics/llm-observability.md>), [experiments](<https://devfeed.tech/topics/experiments.md>)

Tags: [advanced-300](<https://devfeed.tech/tags/advanced-300.md>), [agent](<https://devfeed.tech/tags/agent.md>), [agentic](<https://devfeed.tech/tags/agentic.md>), [agentic-loop](<https://devfeed.tech/tags/agentic-loop.md>), [amazon-athena](<https://devfeed.tech/tags/amazon-athena.md>), [amazon-bedrock](<https://devfeed.tech/tags/amazon-bedrock.md>), [amazon-cloudwatch-logs](<https://devfeed.tech/tags/amazon-cloudwatch-logs.md>), [amazon-s3](<https://devfeed.tech/tags/amazon-s3.md>), [api](<https://devfeed.tech/tags/api.md>), [aws-cost-explorer](<https://devfeed.tech/tags/aws-cost-explorer.md>), [aws-identity-and-access-management-iam](<https://devfeed.tech/tags/aws-identity-and-access-management-iam.md>), [batch](<https://devfeed.tech/tags/batch.md>), [caching](<https://devfeed.tech/tags/caching.md>), [cost](<https://devfeed.tech/tags/cost.md>), [drivers](<https://devfeed.tech/tags/drivers.md>), [experiments](<https://devfeed.tech/tags/experiments.md>), [model](<https://devfeed.tech/tags/model.md>), [optimizing](<https://devfeed.tech/tags/optimizing.md>), [reduce](<https://devfeed.tech/tags/reduce.md>), [roi](<https://devfeed.tech/tags/roi.md>), [routing](<https://devfeed.tech/tags/routing.md>), [technical-how-to](<https://devfeed.tech/tags/technical-how-to.md>)

### AI overview

Compare models by cost per successful task, since token prices alone miss tokenizer differences, verbose output, repeated agent turns, and reasoning tokens. Benchmark with your own prompts: use Amazon Bedrock model evaluation for quality, CountTokens for input comparisons, and invocation `usage` data or logs for token and turn counts. Price cache reads, writes, and uncached input separately; estimates based on published rates may differ from the invoice due to commitments, batch pricing, or provisioned throughput. Reduce cost with prompt caching, routing easy requests to cheaper models, output constraints, and batch inference for bulk single-turn work; batch inference does not support tool calling or multi-turn exchange. To assess ROI, document and measure the human baseline before launch, include human review time in workflow records, and compare agent and control groups on efficiency, quality, cost, and business outcomes. Treat scorecard thresholds as starting points calibrated to your context.

### Source excerpt

Comparing per-token prices hides the real cost of running a model on Amazon Bedrock. This post shows how to calculate the true cost per task across four hidden cost drivers, reduce it with caching, routing, and output constraints, and prove agent ROI with human baselines, holdout experiments, and a five-dimension scorecard.

## From Telecom Expense Management to AI Governance: How IntraTEM Is Expanding TEM

DevFeed: [From Telecom Expense Management to AI Governance: How IntraTEM Is Expanding TEM](<https://devfeed.tech/articles/from-telecom-expense-management-to-ai-governance-how-intratem-is-expanding-tem-130596.md>)

Original publisher: [Read original article](<https://hackernoon.com/from-telecom-expense-management-to-ai-governance-how-intratem-is-expanding-tem?source=rss>)

Author: Sanya Kapoor

Published: 2026-11-12T11:50:12Z

Content type: article

Language: en

Sources: [HackerNoon](<https://devfeed.tech/sources/hackernoon.md>)

Topics: [ai-governance](<https://devfeed.tech/topics/ai-governance.md>), [shadow AI](<https://devfeed.tech/topics/shadow-ai.md>), [SaaS Management](<https://devfeed.tech/topics/saas-management.md>), [cloud cost management](<https://devfeed.tech/topics/cloud-cost-management.md>), [Managed Services](<https://devfeed.tech/topics/managed-services.md>)

Tags: [ai](<https://devfeed.tech/tags/ai.md>), [ai-governance](<https://devfeed.tech/tags/ai-governance.md>), [ai-license-optimization](<https://devfeed.tech/tags/ai-license-optimization.md>), [ai-tool-inventory](<https://devfeed.tech/tags/ai-tool-inventory.md>), [ai-tools](<https://devfeed.tech/tags/ai-tools.md>), [enterprise](<https://devfeed.tech/tags/enterprise.md>), [enterprise-ai-spending](<https://devfeed.tech/tags/enterprise-ai-spending.md>), [good-company](<https://devfeed.tech/tags/good-company.md>), [intratem-ai-governance](<https://devfeed.tech/tags/intratem-ai-governance.md>), [saas](<https://devfeed.tech/tags/saas.md>), [saas-expense-management](<https://devfeed.tech/tags/saas-expense-management.md>), [shadow-ai-detection](<https://devfeed.tech/tags/shadow-ai-detection.md>), [shadow-ai-management](<https://devfeed.tech/tags/shadow-ai-management.md>), [spending](<https://devfeed.tech/tags/spending.md>), [telecom](<https://devfeed.tech/tags/telecom.md>)

### AI overview

IntraTEM is extending its telecom and SaaS expense audit model to AI governance, focusing on identifying AI features and seats hidden in approved software licenses as well as unauthorized tools. The article cites findings that 43% of organizations cannot inventory or audit workforce AI tools, 73% of SaaS vendors charge separately for AI features, and 77% of IT leaders have found AI applications operating without IT awareness. IntraTEM frames this as both a visibility and cost-control problem: its existing license-versus-usage work can surface overlooked AI spending, while its fully managed, human-led service follows through on audit findings. The company says it is not policing employee AI use; its focus is ensuring seats appear on reviewed invoices and someone is accountable for next steps. The article distinguishes this approach from software-only TEM platforms and security tools that do not address budgets.

### Source excerpt

Discover how IntraTEM is extending its telecom and SaaS expense management expertise into AI governance to uncover hidden tools, licenses, & enterprise spending

## Automate PRs to close the loop on cloud waste

DevFeed: [Automate PRs to close the loop on cloud waste](<https://devfeed.tech/articles/automate-prs-to-close-the-loop-on-cloud-waste-130791.md>)

Original publisher: [Read original article](<https://www.harness.io/blog/close-the-loop-on-cloud-waste>)

Author: Uri Scheiner Rohit Reddy Kaliki

Published: 2026-11-12T00:00:00Z

Content type: release

Language: en

Sources: [Harness](<https://devfeed.tech/sources/harness-blog.md>)

Topics: [cloud cost management](<https://devfeed.tech/topics/cloud-cost-management.md>), [human in the loop agent](<https://devfeed.tech/topics/human-in-the-loop-agent.md>), [internal developer platform automation](<https://devfeed.tech/topics/internal-developer-platform-automation.md>)

Tags: [agentic](<https://devfeed.tech/tags/agentic.md>), [agents](<https://devfeed.tech/tags/agents.md>), [ai](<https://devfeed.tech/tags/ai.md>), [ai-agent](<https://devfeed.tech/tags/ai-agent.md>), [automate](<https://devfeed.tech/tags/automate.md>), [automation](<https://devfeed.tech/tags/automation.md>), [backlog](<https://devfeed.tech/tags/backlog.md>), [bill](<https://devfeed.tech/tags/bill.md>), [blast-radius](<https://devfeed.tech/tags/blast-radius.md>), [change](<https://devfeed.tech/tags/change.md>), [changes](<https://devfeed.tech/tags/changes.md>), [citizen-developers](<https://devfeed.tech/tags/citizen-developers.md>), [cloud](<https://devfeed.tech/tags/cloud.md>), [cloud-costs](<https://devfeed.tech/tags/cloud-costs.md>), [cost](<https://devfeed.tech/tags/cost.md>), [harness](<https://devfeed.tech/tags/harness.md>), [iac](<https://devfeed.tech/tags/iac.md>), [pull-requests](<https://devfeed.tech/tags/pull-requests.md>), [recommendations](<https://devfeed.tech/tags/recommendations.md>), [waste](<https://devfeed.tech/tags/waste.md>)

### AI overview

Harness's Infrastructure Efficiency Loop connects cloud cost and drift findings to infrastructure code changes through reviewable pull requests. Infrastructure Deployments correlates provisioned resources with cost data, showing costs after discounts and enterprise agreements alongside optimization opportunities. The Infrastructure Remediation Agent uses workspace context to edit code and open a PR; teams can configure its pipeline and exclude workspaces, and people still review and merge every change. Infrastructure Blast Radius scores planned changes using affected resources and dependencies, explains the score, and exposes risk and cost in the pipeline so teams can set approval gates. It currently assesses a single workspace; cross-workspace dependencies are planned. After deployment, cost monitoring can surface the next opportunity.

### Source excerpt

Finding cloud waste is half the job. Harness agents turn cost recommendations into reviewable IaC pull requests and track what gets applied. | Blog

[Next page](<https://devfeed.tech/topics/cloud-cost-management.md?cursor=WyIyMDI2LTExLTEyVDAwOjAwOjAwKzAwOjAwIiwgIjg1MjVlNWFkLTkxYmMtNDc2NC1hM2FmLTA5YTU3YWI5MDVhMyJd>)