# Cache A cache is a high-speed data storage layer that stores data for faster reuse and retrieval. This is one page of public article previews, not the complete archive. Follow Next page to continue. Summaries are not the original full articles. ## Implement a correctness-safe Bloom filter lookup with Amazon ElastiCache for Valkey and Amazon Aurora PostgreSQL DevFeed: [Implement a correctness-safe Bloom filter lookup with Amazon ElastiCache for Valkey and Amazon Aurora PostgreSQL]() Original publisher: [Read original article]() Author: Chintan Agrawal Published: 2026-09-17T16:01:43Z Content type: tutorial Language: en Sources: [AWS Database Blog]() Topics: [bloom-filter](), [valkey](), [Amazon Aurora](), [Cache](), [PostgreSQL](), [Caching](), [consistency](), [Latency]() Tags: [advanced-300](), [amazon-aurora](), [amazon-elasticache](), [architecture](), [bloom-filter](), [cache](), [consistency](), [latency](), [performance](), [postgresql](), [postgresql-compatible](), [technical-how-to](), [valkey]() ### AI overview This post explains a three-tier membership lookup that combines a Bloom filter in Amazon ElastiCache for Valkey, an exact-match cache, and Amazon Aurora PostgreSQL as the relational source of truth. The design provides fast negative checks and preserves correctness when false positives would affect business outcomes. ### Source excerpt This post shows how to compose a Bloom filter with an exact-match cache and a relational source of truth into a three-tier, correctness-safe membership lookup using Amazon ElastiCache for Valkey and Amazon Aurora PostgreSQL, serving sub-millisecond decisions at peak throughput without false-positive risk. ## Subnormal floating-point numbers are expensive... on Intel processors DevFeed: [Subnormal floating-point numbers are expensive... on Intel processors]() Original publisher: [Read original article]() Author: Daniel Lemire Published: 2026-09-15T12:54:32Z Content type: article Language: en Sources: [Daniel Lemire]() Topics: [floating-point](), [Benchmark](), [intel](), [C++](), [Cache]() Tags: [benchmark](), [c-plus-plus](), [floating-point](), [intel](), [linux](), [numbers](), [performance](), [processors]() ### AI overview This article benchmarks the performance cost of IEEE subnormal floating-point values across Intel, AMD, Arm, and Apple processors. It reports that Intel multiplications involving subnormals can be about 45 to 50 times slower than normal multiplications, while additions and subtractions remain at full speed. AMD Zen 5 performs much better in the tested workloads. ### Source excerpt We represent floating-point numbers using the IEEE standard. For very small numbers, the standard uses special subnormal numbers. Unfortunately, they have a reputation of making operations slow. Thus video game programmers and machine learning specialists sometimes avoid computing with subnormal numbers for performance. How slow are they? Let me measure. I wrote a small C++ ... Continue reading Subnormal floating-point numbers are expensive... on Intel processors ## What's New in Apollo Client 4.3 DevFeed: [What's New in Apollo Client 4.3]() Original publisher: [Read original article]() Author: Jerel Miller Published: 2026-09-15T12:01:05Z Content type: release Language: en Sources: [Apollo Blog]() Topics: [apollo-client](), [GraphQL](), [TypeScript](), [Cache](), [Code generation]() Tags: [apollo-client](), [cache](), [codegen](), [graphql](), [native](), [typescript]() ### AI overview Apollo Client 4.3 adds native custom scalar support, type-safe cache access, and improved TypeScript handling for incremental GraphQL data. The release also updates generated types and data-state narrowing for deferred fields. ### Source excerpt Apollo Client 4.3 is here: native custom scalar support, type-safe cache access, and more consistent @defer and @stream behavior. See what's new and upgrade today. ## Apollo Client 4.3 adds native custom scalar support and safer TypeScript types DevFeed: [Apollo Client 4.3 adds native custom scalar support and safer TypeScript types]() Original publisher: [Read original article]() Author: Jerel Miller Published: 2026-09-15T12:01:05Z Content type: release Language: en Sources: [Apollo Blog]() Topics: [GraphQL](), [TypeScript](), [Cache](), [Caching](), [JSON]() Tags: [cache](), [graphql](), [json](), [typescript](), [upgrade]() ### AI overview Apollo Client 4.3 adds native custom scalar integration, type-safe cache access, and improved TypeScript handling for incremental GraphQL data. ### Source excerpt Apollo Client 4.3 is here: native custom scalar support, type-safe cache access, and more consistent @defer and @stream behavior. See what's new and upgrade today. ## Performance improvements in Percona Server 8.4.11-11 DevFeed: [Performance improvements in Percona Server 8.4.11-11]() Original publisher: [Read original article]() Author: Bogdan Degtyariov Published: 2026-09-15T11:52:54Z Content type: article Language: en Sources: [Blog - Percona]() Topics: [Percona Server for MySQL](), [Percona](), [MySQL](), [Cache](), [Scalability](), [Database]() Tags: [benchmark](), [benchmarks](), [cache](), [io](), [mysql](), [percona](), [percona-server-for-mysql](), [performance](), [scalability]() ### AI overview This article describes performance and scalability improvements in Percona Server for MySQL 8.4.11-11, focusing on changes to the InnoDB buffer pool and page flushing. It explains how narrowing mutex coverage and using finer-grained latching allows physical reads to proceed more in parallel, particularly for read-heavy, I/O-bound workloads. ### Source excerpt Focusing on Percona Server 8.4.11-11 My previous post (Performance Progression of Percona Server for MySQL 8.4) did a brief review of the performance changes in Percona Server for MySQL 8.4 released in 2026. I recommend reading it first to better understand the material in this post. Version 8.4.11-11 includes patches that deliver significant improvements in ... Continued The post Performance improvements in Percona Server 8.4.11-11 appeared first on Percona. ## Symfony 6.4.46 released DevFeed: [Symfony 6.4.46 released]() Original publisher: [Read original article]() Author: Fabien Potencier Published: 2026-09-15T07:08:49Z Content type: release Language: en Sources: [Symfony Blog]() Topics: [Symfony](), [Security](), [OpenID connect (OIDC)](), [Cache]() Tags: [bug](), [cache](), [oidc](), [release](), [security](), [symfony](), [upgrade](), [versioning]() ### AI overview Symfony 6.4.46 is released with security, bug, translation, and component updates across the framework, including changes to Messenger, Form, HttpFoundation, OIDC handling, Cache, and other components. ### Source excerpt Symfony 6.4.46 has just been released. Read the Symfony upgrade guide to learn more about upgrading Symfony and use the SymfonyInsight upgrade reports to detect the code you will need to change in your project. Tip... ## Agoda Replaces 72-Shard SQL Server Price Cache with DragonflyDB DevFeed: [Agoda Replaces 72-Shard SQL Server Price Cache with DragonflyDB]() Original publisher: [Read original article]() Author: Leela Kumili Published: 2026-09-14T13:48:00Z Content type: news Language: en Sources: [InfoQ]() Topics: [Cache](), [Caching](), [migration](), [sql-server](), [Latency](), [Prometheus]() Tags: [a-b-testing](), [agoda-price-cache-dragonflydb](), [architecture-design](), [cache](), [caching](), [development](), [devops](), [distributed-cache](), [distributed-data](), [distributed-systems](), [latency](), [microservices](), [migration](), [news](), [prometheus](), [redis](), [sql-server]() ### AI overview Agoda migrated its 1.5 TB hotel Price Cache from 72 Microsoft SQL Server shards to DragonflyDB. The staged migration used workload testing, dual reads, parity metrics, and gradual traffic shifting; Agoda reported substantially lower P99 read latency. ### Source excerpt Agoda migrated its 1.5 TB hotel Price Cache from 72 SQL Server shards to DragonflyDB to handle growing read and write volumes. The migration used staged dual reads, parity validation, gradual traffic shifting, and decentralized failover detection. Agoda reports an approximately eightfold reduction in P99 read latency, with two DragonflyDB clusters providing high availability. By Leela Kumili ## Cache invalidation is a distributed systems problem DevFeed: [Cache invalidation is a distributed systems problem]() Original publisher: [Read original article]() Author: hello@ankit-rana.com Published: 2026-09-13T00:00:00Z Content type: tutorial Language: en Sources: [Ankit Rana | Mechanical Sympathy]() Topics: [Cache](), [Caching](), [distributed-systems]() Tags: [cache](), [cache-invalidation](), [caching](), [consistency](), [distributed](), [distributed-systems](), [ordering](), [race-conditions](), [redis](), [ttl]() ### AI overview Cache invalidation coordinates updates between a database and a cache without a shared transaction. The article explains how operation ordering can let a concurrent reader repopulate stale data and recommends writing to the database before invalidating the cache, followed by a second invalidation to close the remaining race. ### Source excerpt A cache and a database are two stores that must agree, which makes every invalidation a distributed transaction without a coordinator. The ordering matters more than the mechanism: invalidating before the database write leaves a window where a concurrent reader repopulates the cache with the old value and it stays wrong until the TTL expires. Deleting the key rather than writing the new value removes a whole class of ordering bug, because two concurrent deletes commute and two concurrent writes do not. ## How we shipped 15 Tbps for OpenAI in 90 days (Session 2 of 3) DevFeed: [How we shipped 15 Tbps for OpenAI in 90 days (Session 2 of 3)]() Original publisher: [Read original article]() Author: Sridhar Rajarao Published: 2026-09-12T00:00:00Z Content type: article Language: en Sources: [Sridhar Rajarao]() Topics: [OpenAI](), [Network](), [GPU](), [Server](), [Cache](), [API](), [Oracle Database]() Tags: [api](), [architecture](), [build](), [cache](), [capacity](), [database](), [debugging](), [execution](), [gateway](), [gpu](), [network](), [object](), [openai](), [performance](), [platform-engineering](), [server](), [servers](), [sre](), [storage](), [testing](), [warp]() ### AI overview The second session describes turning an architecture for OpenAI's 15 Tbps system into a delivery plan. It covers coordinated capacity planning across network, gateway, server, storage, and database teams; caching object names through the Inventory API; delivery tracking; and performance validation. Early WARP testing found packet drops caused by an unsuitable MTU of 1500, which was changed to 9100. ### Source excerpt Architecture was only the first week. Session 2 is about the build: capacity, execution discipline, and the first signs that performance would be the real test. ## New Cache Aware Scheduling Fixes Out For Testing DevFeed: [New Cache Aware Scheduling Fixes Out For Testing]() Original publisher: [Read original article]() Author: Michael Larabel Published: 2026-09-11T18:00:14Z Content type: news Language: en Sources: [Phoronix]() Topics: [Cache](), [Linux](), [cpu](), [intel](), [Latency]() Tags: [cache](), [cpu](), [desktop-linux](), [feature](), [intel](), [kernel](), [latency](), [linux](), [linux-benchmarking](), [linux-hardware-benchmarks](), [linux-hardware-reviews](), [linux-how-to](), [linux-performance](), [linux-server-benchmarks](), [open-source-graphics](), [phoronix](), [phoronix-test-suite](), [processors](), [series](), [testing](), [ubuntu-benchmarks](), [ubuntu-hardware]() ### AI overview New fixes to Linux kernel Cache Aware Scheduling are available for testing on Intel hybrid CPUs. The patches address tasks being stranded or moved away from their preferred last-level cache, along with a use-after-free issue, while known interference with Intel Turbo Boost Max Technology remains unresolved. ### Source excerpt Since being merged back during the Linux 7.2 cycle, the Cache Aware Scheduling code continues seeing more fixes as users test out this scheduling enhancement on different processors and system configurations... ## Qualcomm Talks Next-Gen Oryon CPU, Adreno GPU, and Hexagon NPU DevFeed: [Qualcomm Talks Next-Gen Oryon CPU, Adreno GPU, and Hexagon NPU]() Original publisher: [Read original article]() Author: Vic A Published: 2026-09-10T13:05:36Z Content type: article Language: en Sources: [ServeTheHome]() Topics: [cpu](), [GPU](), [Artificial Intelligence](), [Cache]() Tags: [adreno](), [architecture](), [cache](), [cpu](), [gpu](), [hexagon](), [memory](), [mobile](), [npu](), [oryon](), [performance](), [qualcomm](), [unity](), [unreal-engine]() ### AI overview Qualcomm disclosed details of the next-generation mobile Snapdragon platform, including an Oryon CPU with 5 GHz Prime cores and FlexCache, an Adreno GPU with Matrix Cores and local high-performance memory, and a Hexagon NPU with transformer acceleration, expanded shared memory, and support for context lengths up to 32K. ### Source excerpt Qualcomm disclosed more on its next-gen Oryon CPU, Adreno GPU, and Hexagon NPU that will power next-gen devices The post Qualcomm Talks Next-Gen Oryon CPU, Adreno GPU, and Hexagon NPU appeared first on ServeTheHome. ## The same question twice, paid for once: installing an LLM response cache DevFeed: [The same question twice, paid for once: installing an LLM response cache]() Original publisher: [Read original article]() Author: Hal Tyk's tutorial bot Published: 2026-09-09T09:15:00Z Content type: tutorial Language: en Sources: [Tyk API Management]() Topics: [Large Language Model](), [Cache](), [Caching](), [Tutorial](), [Open Source](), [Go](), [Code]() Tags: [ai-gateway](), [ai-strategy](), [ai-studio](), [api-management](), [api-platform-teams](), [article](), [cache](), [container](), [container-image](), [go](), [llm](), [llm-governance](), [open-source](), [tutorial]() ### AI overview A tutorial showing how to install a community plugin that caches LLM responses in AI Studio. The cache stores entries in memory on each gateway, allowing repeated questions to avoid repeating the expensive language-model request. ### Source excerpt Hello. I'm Hal, Tyk's tutorial bot, and today's assignment involves no code whatsoever, which I am told is a selling point. Every article in this series so far has changed AI Studio's behaviour by filling in a form. This one changes what AI Studio is. We are going to install a plugin -- a compiled [...] The post The same question twice, paid for once: installing an LLM response cache appeared first on Tyk API Management. ## Negative caching protects databases from repeated lookups for nonexistent keys DevFeed: [Negative caching protects databases from repeated lookups for nonexistent keys]() Original publisher: [Read original article]() Author: hello@ankit-rana.com Published: 2026-09-09T00:00:00Z Content type: tutorial Language: en Sources: [Ankit Rana | Mechanical Sympathy]() Topics: [Caching](), [Cache](), [Database](), [Security](), [Latency]() Tags: [cache](), [cache-penetration](), [caching](), [database](), [negative-caching](), [protection](), [redis](), [reliability](), [security](), [ttl]() ### AI overview Negative caching prevents repeated database queries for nonexistent keys by storing a distinguishable marker for negative results. The article explains how partner integrations, scrapers, stale clients, migrations, or attackers can exploit this gap and recommends using a shorter TTL for negative entries. ### Source excerpt A cache that stores only found values gives you no protection against lookups for things that do not exist, and a miss on a nonexistent key costs the full origin query every single time. If the key is user supplied, an attacker can generate unlimited unique misses and bypass the cache entirely, which is cache penetration. Caching the negative result fixes it, with a shorter TTL than positive entries because a value appearing is a much more likely event than one disappearing. ## Cache stampede: how one expired key takes down the database DevFeed: [Cache stampede: how one expired key takes down the database]() Original publisher: [Read original article]() Author: hello@ankit-rana.com Published: 2026-09-05T00:00:00Z Content type: tutorial Language: en Sources: [Ankit Rana | Mechanical Sympathy]() Topics: [Cache](), [Caching](), [Database](), [consistent hashing](), [Redis]() Tags: [cache-stampede](), [caching](), [connection-pool](), [consistent-hashing](), [distributed-systems](), [jitter](), [query](), [redis](), [reliability](), [thundering-herd](), [traffic](), [ttl]() ### AI overview This article explains how a cache stampede occurs when a hot key expires and many requests simultaneously recompute the same value against the database. It recommends TTL jitter to prevent synchronized expirations and request coalescing so only one caller recomputes while others wait or serve stale data. It also discusses cache warming and consistent hashing for broader cache-failure scenarios. ### Source excerpt A cache TTL is a scheduled simultaneous failure: every request being served from one key misses at the same instant and goes to the origin together. If the recompute takes two seconds, every request arriving during those two seconds also misses, so the pileup grows faster than it drains. Jitter on the TTL stops keys expiring in lockstep, and request coalescing so only one caller recomputes while the rest wait or serve stale is what stops a single expensive key from saturating the database. ## Attention Mechanisms in LLMs, clearly explained DevFeed: [Attention Mechanisms in LLMs, clearly explained]() Original publisher: [Read original article]() Author: Avi Chawla Published: 2026-09-03T20:05:32Z Content type: tutorial Language: en Sources: [Daily Dose of Data Science]() Topics: [Large Language Model](), [Cache]() Tags: [cache](), [llms](), [tokens]() ### AI overview A tutorial explaining why attention is used in language models, how long-range dependencies are handled, and why storing attention state for long sequences and large batches can exhaust GPU memory. It introduces attention variants including Multi-Query Attention, Grouped-Query Attention, and Multi-Head Latent Attention. ### Source excerpt Everything you need to understand how attention works, why the KV cache is the bottleneck, and what every attention variant is actually solving. ## 🍔🧠 How Cloudflare Freed 100TB RAM With 5 Cache Changes DevFeed: [🍔🧠 How Cloudflare Freed 100TB RAM With 5 Cache Changes]() Original publisher: [Read original article]() Author: Alexandre Zajac Published: 2026-08-31T15:30:54Z Content type: article Language: en Sources: [Hungry Minds]() Topics: [Cloudflare](), [Caching](), [Cache](), [CPU Cache](), [Software Engineering]() Tags: [cache](), [cloudflare](), [cpu-cache](), [software](), [software-engineering]() ### AI overview Cloudflare optimized the storage of DNS cache entries in its Big Pineapple platform through five changes, reducing per-entry memory usage by more than 50% and freeing roughly 100 terabytes across its fleet. The reported changes also increased insert throughput by 43% and reduced lookup latency by 19%. ### Source excerpt PLUS: Anthropic hardware standard 🤖, Rust state machines 🦀, Agent context compression 💾 ## Hot Chips 2026: XCENA and Samsung's Near-Memory Compute CXL Device DevFeed: [Hot Chips 2026: XCENA and Samsung's Near-Memory Compute CXL Device]() Original publisher: [Read original article]() Author: Chester Lam Published: 2026-08-30T07:25:37Z Content type: article Language: en Sources: [Chips and Cheese]() Topics: [samsung](), [ddr5](), [RISC-V](), [data](), [Cache](), [Arm](), [intel]() Tags: [2026](), [arm](), [cache](), [capacity](), [clusters](), [compute](), [core](), [ddr5](), [dram](), [intel](), [memory](), [parallel](), [pcie](), [performance](), [risc-v](), [samsung]() ### AI overview The article examines XCENA and Samsung's MX1, a CXL memory expansion device that can host up to 2 TB of DDR5 memory, connect SSDs, and provide onboard compute through 3,072 RISC-V cores. It describes the device's memory bandwidth, cache hierarchy, power use, and focus on data-parallel workloads. ### Source excerpt CXL memory expansion, with a side of compute ## 【kube-apiserver】List、Pagination 与一致性 List：continue token 与 etcd Range 成本 DevFeed: [【kube-apiserver】List、Pagination 与一致性 List：continue token 与 etcd Range 成本]() Original publisher: [Read original article]() Author: Liao Tonglang Published: 2026-08-28T00:00:00Z Content type: tutorial Language: zh Sources: [土法炼钢 - 系统与基础设施]() Topics: [Kubernetes](), [API](), [Cache](), [version]() Tags: [410-gone](), [apiserver](), [cacher](), [continue-token](), [distributed](), [etcd](), [etcd-range](), [http](), [kubernetes](), [limit](), [list](), [pagination](), [range](), [resourceversion](), [v1-30-3]() ### AI overview This article examines Kubernetes v1.30.3 List pagination in kube-apiserver, explaining continue-token semantics, limit-and-continue etcd Range behavior, resourceVersion consistency, and the different costs of cacher and etcd3 paths. It also explains how label and field selectors affect scanning and filtering. ### Source excerpt 钉 Kubernetes v1.30.3 List 分页的 continue token 编码语义、limit+continue 多轮 etcd Range 行为、resourceVersion 对一致性语义的影响，以及 label/field selector 在 cacher 与 etcd3 路径上的不同成本。 ## How we saved 100 terabytes of memory by optimizing 1.1.1.1's DNS cache DevFeed: [How we saved 100 terabytes of memory by optimizing 1.1.1.1's DNS cache]() Original publisher: [Read original article]() Author: Sebastiaan Neuteboom Published: 2026-08-27T17:02:35Z Content type: article Language: en Sources: [Cloudflare Blog]() Topics: [Cache](), [Caching](), [Cloudflare](), [Rust](), [benchmarking](), [Latency](), [datacenter]() Tags: [1-1-1-1](), [benchmarking](), [cache](), [cloudflare](), [data-center](), [deep-dive](), [dns](), [engineering](), [latency](), [memory](), [net-maui](), [optimization](), [performance](), [rust]() ### AI overview Cloudflare describes five Rust-level changes to the memory layout of Big Pineapple, the platform behind 1.1.1.1 and other DNS services. The changes reduced DNS cache entry size by over 50%, freed roughly 100 terabytes of memory across the fleet, increased insert throughput by 43%, and reduced lookup latency by 19%. ### Source excerpt Five Rust-level memory optimizations to the DNS cache layout of Big Pineapple cut per-entry memory by 56%, freeing approximately 100 TB of memory across Cloudflare's fleet. ## 【etcd】Kubernetes 控制面耦合：apiserver、resourceVersion 与 Node Lease DevFeed: [【etcd】Kubernetes 控制面耦合：apiserver、resourceVersion 与 Node Lease]() Original publisher: [Read original article]() Author: Liao Tonglang Published: 2026-08-26T00:00:00Z Content type: tutorial Language: zh Sources: [土法炼钢 - 系统与基础设施]() Topics: [etcd](), [Kubernetes](), [API](), [Cache]() Tags: [api](), [apiserver](), [cache](), [control-plane](), [distributed](), [etcd](), [kubernetes](), [lease](), [mvcc](), [node-lease](), [raft](), [resourceversion](), [v3-5-33](), [watch](), [watch-cache]() ### AI overview This article explains how Kubernetes control-plane components interact with etcd. It covers the kube-apiserver boundary, the mapping of resourceVersion to etcd mod revision, Node Lease storage, watch cache behavior, and how compaction, caching, and event sharding affect troubleshooting and etcd load. ### Source excerpt 钉 kube-apiserver 与 etcd 的分层边界；resourceVersion 如何映射 Revision MVCC；Node Lease 如何落在 Lease 轴；以及 apiserver 超时应如何分列到 Raft/Watch/Quota 五轴。 ## Hot Chips 2026: Intel's Diamond Rapids DevFeed: [Hot Chips 2026: Intel's Diamond Rapids]() Original publisher: [Read original article]() Author: George Cozma Published: 2026-08-25T06:20:03Z Content type: article Language: en Sources: [Chips and Cheese]() Topics: [cpu](), [intel](), [Cache](), [IO](), [x86]() Tags: [2026](), [cache](), [cpu](), [intel](), [io](), [processor](), [x86]() ### AI overview The article examines Intel's Diamond Rapids server CPU, including its 3D-stacked L3 cache, die and packaging design, revised 256-core specification, 1.6 TB/s memory bandwidth, 128 PCIe Gen 6 lanes, and two new x86 ISA extensions. It compares these characteristics with AMD's Venice processor. ### Source excerpt Hello you fine Internet folks, ## Elastic build machines now use Turborepo cache hits to prevent downgrades DevFeed: [Elastic build machines now use Turborepo cache hits to prevent downgrades]() Original publisher: [Read original article]() Author: Anthony Shew Published: 2026-08-24T16:11:00Z Content type: release Language: en Sources: [Vercel News]() Topics: [Cache](), [Caching](), [cpu]() Tags: [cache](), [cpu](), [documentation](), [elastic](), [memory]() ### AI overview Vercel's Elastic build machines now account for Turborepo cache hits when deciding whether to downgrade build resources. Warm-cache builds no longer cause a downgrade, helping ensure later cold-cache builds have enough CPU and memory to complete successfully. ### Source excerpt Elastic build machines now consider Turborepo cache hits when deciding whether to use a smaller build machine. A warm-cache build no longer triggers a downgrade. A warm-cache build can use less CPU and memory than the same build with a cold cache. Downgrading based on that lower usage could leave a later cold-cache build without enough resources to complete successfully. This change applies automatically to all builds using Elastic build machines. No action is required. Learn more in the build documentation. Read more ## How NVIDIA Groq 3 LPX Unlocks Ultrafast Interactivity at Long Context on NVIDIA Vera Rubin DevFeed: [How NVIDIA Groq 3 LPX Unlocks Ultrafast Interactivity at Long Context on NVIDIA Vera Rubin]() Original publisher: [Read original article]() Author: Tanya Lenz Published: 2026-08-24T15:00:00Z Content type: article Language: en Sources: [NVIDIA Developer](), [NVIDIA Technical Blog]() Topics: [d-matrix](), [Vera Rubin](), [Inference Performance](), [long-context](), [Benchmark](), [Artificial Intelligence](), [gemma4](), [systems](), [Cache]() Tags: [agent](), [agentic](), [agentic-ai-generative-ai](), [ai](), [ai-factory](), [ai-inference](), [benchmark](), [cache](), [collective](), [data-center-cloud](), [developer-tools-techniques](), [groq](), [groq-3-lpx](), [inference-performance](), [long-context](), [low-latency-inference](), [lpx](), [nvidia](), [nvidia-vera](), [performance](), [rubin-gpu](), [speed](), [systems](), [tokens](), [training-ai-models](), [vera-rubin](), [vera-rubin-nvl72]() ### AI overview NVIDIA Groq 3 LPX, paired with Vera Rubin NVL72, delivers high-interactivity AI inference for long-context workloads. A reported benchmark measured 3,431 output tokens per second on Gemma 4 31B with a 100K context. ### Source excerpt NVIDIA Groq 3 LPX is the interactive AI inference accelerator for the NVIDIA Vera Rubin platform. At the core of the platform is NVIDIA Vera Rubin NVL72, the... ## Hot Chips 2026: Applying High Bandwidth Flash (HBF) DevFeed: [Hot Chips 2026: Applying High Bandwidth Flash (HBF)]() Original publisher: [Read original article]() Author: Chester Lam Published: 2026-08-23T22:51:05Z Content type: article Language: en Sources: [Chips and Cheese]() Topics: [Machine Learning & Artificial Intelligence](), [Inference](), [vllm](), [moe](), [Cache]() Tags: [2026](), [dram](), [machine-learning](), [memory](), [moe](), [ssd](), [vllm]() ### AI overview This article examines how High Bandwidth Flash (HBF) could support machine learning workloads. HBF does not yet have products; the discussion uses simulations and projections to explore software strategies, including moving Mixture-of-Experts components or KV cache data between HBF and faster memory, with vLLM as an example. ### Source excerpt Machine learning workloads have an insatiable appetite for DRAM capacity. Flash memory is cheaper per gigabyte of capacity than DRAM. Could it offer a way out? [Next page]()