正在加载视频...

视频加载失败

加载此视频时出现问题。这可能是由于临时网络问题，或视频可能不可用。

You asked for it, so here it is. Visualizing CPU cache speeds relative to RAM. Cache optimization is important too!

Ben Dicken

51,121 subscribers

1,474,881 次观看 • 2 年前 •via X (Twitter)

Anya Rossi• Live Now

Private livecam show

0 条评论

暂无评论

原始帖子的评论将显示在这里

相关视频

Some of the biggest news today is Cloudflare Worker's Cache. So I built this small catalog to experience it. It allows you to purge the cache for certain pages, or the entire catalog. It also shows you the exact caching headers used. See it in action or try it yourself:

Some of the biggest news today is Cloudflare Worker's Cache. So I built this small catalog to experience it. It allows you to purge the cache for certain pages, or the entire catalog. It also shows you the exact caching headers used. See it in action or try it yourself:

Jilles Soeters

17,920 次观看 • 3 个月前

Submitted Easy Vertex Anim for #UE5.8 too. You can create pivot painter 2 effects and bake chaos cache to vat, 100% NO AI, check it out!

Submitted Easy Vertex Anim for #UE5.8 too. You can create pivot painter 2 effects and bake chaos cache to vat, 100% NO AI, check it out!

Inu Games

10,373 次观看 • 4 个月前

MeidasTouch It was still in my cache

MeidasTouch It was still in my cache

DANNYonPC

11,097 次观看 • 2 个月前

WanGP's latest MiniMax H3 update is huge: - Way faster Generation - 2X faster Ref2VA - Lower-RAM & new accelerators - Long videos with smoother visual/audio transitions. - Huge speed gains with Sol-Attention, First Block Cache, etc. And Pinokio launcher is optimized for it 👇

WanGP's latest MiniMax H3 update is huge: - Way faster Generation - 2X faster Ref2VA - Lower-RAM & new accelerators - Long videos with smoother visual/audio transitions. - Huge speed gains with Sol-Attention, First Block Cache, etc. And Pinokio launcher is optimized for it 👇

cocktail peanut

61,351 次观看 • 2 个月前

A lot of people asked me to post FOOTage of Iuno's burst It features her, you guessed it, feet So here it is Enjoy!

Sensitive content

A lot of people asked me to post FOOTage of Iuno's burst It features her, you guessed it, feet So here it is Enjoy!

Geralt of Goonia

261,127 次观看 • 6 个月前

cache is out so use this easy b site utility and win the round (hopefully)

cache is out so use this easy b site utility and win the round (hopefully)

mirage instant window smoke dritter spawn

102,149 次观看 • 6 个月前

Want the latest, flashiest skins? Lucky Cache is here for you every week! Every Friday to Sunday, Step into the Lucky Cache and challenge destiny! (level 80 needed) 🎁 One free draw every week A single draw could change everything and rewards can refresh 3 times a week — are you ready? Luck doesn't wait!

Want the latest, flashiest skins? Lucky Cache is here for you every week! Every Friday to Sunday, Step into the Lucky Cache and challenge destiny! (level 80 needed) 🎁 One free draw every week A single draw could change everything and rewards can refresh 3 times a week — are you ready? Luck doesn't wait!

Call of Odin's Chosen

121,198 次观看 • 10 个月前

cache-cache à paris

cache-cache à paris

doodles

126,930 次观看 • 1 年前

Day 11/90 of Inference Engineering How does vLLM work and how is it used in production? Before we discuss how vLLM works internally, it helps to understand what vLLM is. At a high level, vLLM is an inference engine that is designed to serve LLMs to thousands of concurrent users efficiently while managing scarce compute and memory. The goal for vLLM is to maximize throughput and minimize latency; optimizing for the best inference economics and experience for end users. With every request from the end user, it eventually ends up in the engine core, gets scheduled alongside other requests from other concurrent users, executes on the GPU, and updates the KV cache with the new key and value vectors, and streams the tokens back to the user. The Scheduler decides what requests should execute next while continuously batching requests together to maximize GPU utilization. Continuous batching is an inference optimization that allows new requests to join a running batch as other requests finish generating tokens. This helps with keeping the GPU utilization high instead of letting it sit idle waiting for an entire batch to complete generating. After the scheduler dispatches the selected batch to the Model Executor, the Model Executor prepares the tensors and metadata required for inference, retrieves each request’s block table from KV Cache Manager, launches the optimized transformer forward pass on the GPU, computes the logits, updates the KV cache with the new key and value vectors, and finally returns the results for sampling and streaming. The KV Cache Manager uses the PagedAttention memory layout to allocate fixed-size cache blocks on demand and maintains a Free Block Queue on the CPU that tracks which blocks in the GPU’s Paged KV Cache are currently free. When a request needs additional KV cache space, the KV Cache manager takes a free block from the queue and assigns it to that request, thus avoiding an expensive search through GPU memory for available cache blocks. All of these components form the core of vLLM’s inference engine. The Scheduler determines what requests are executed, the Model Executor determines how those requests are executed, the KV Cache Manager determines where each request’s KV cache lives using the PagedAttention Memory Layout. This architecture enables vLLM to serve thousands of concurrent requests with high throughput, low latency, and efficient GPU memory utilization. Heres a little animation that visualizes everything! - I've also completed the forward pass for my mnist.c project. I had a nice chat with shrey birmiwal, such a knowledgeable guy. Excited to learn more about vLLM and implement a tiny-vLLM one day.

Day 11/90 of Inference Engineering How does vLLM work and how is it used in production? Before we discuss how vLLM works internally, it helps to understand what vLLM is. At a high level, vLLM is an inference engine that is designed to serve LLMs to thousands of concurrent users efficiently while managing scarce compute and memory. The goal for vLLM is to maximize throughput and minimize latency; optimizing for the best inference economics and experience for end users. With every request from the end user, it eventually ends up in the engine core, gets scheduled alongside other requests from other concurrent users, executes on the GPU, and updates the KV cache with the new key and value vectors, and streams the tokens back to the user. The Scheduler decides what requests should execute next while continuously batching requests together to maximize GPU utilization. Continuous batching is an inference optimization that allows new requests to join a running batch as other requests finish generating tokens. This helps with keeping the GPU utilization high instead of letting it sit idle waiting for an entire batch to complete generating. After the scheduler dispatches the selected batch to the Model Executor, the Model Executor prepares the tensors and metadata required for inference, retrieves each request’s block table from KV Cache Manager, launches the optimized transformer forward pass on the GPU, computes the logits, updates the KV cache with the new key and value vectors, and finally returns the results for sampling and streaming. The KV Cache Manager uses the PagedAttention memory layout to allocate fixed-size cache blocks on demand and maintains a Free Block Queue on the CPU that tracks which blocks in the GPU’s Paged KV Cache are currently free. When a request needs additional KV cache space, the KV Cache manager takes a free block from the queue and assigns it to that request, thus avoiding an expensive search through GPU memory for available cache blocks. All of these components form the core of vLLM’s inference engine. The Scheduler determines what requests are executed, the Model Executor determines how those requests are executed, the KV Cache Manager determines where each request’s KV cache lives using the PagedAttention Memory Layout. This architecture enables vLLM to serve thousands of concurrent requests with high throughput, low latency, and efficient GPU memory utilization. Heres a little animation that visualizes everything! - I've also completed the forward pass for my mnist.c project. I had a nice chat with shrey birmiwal, such a knowledgeable guy. Excited to learn more about vLLM and implement a tiny-vLLM one day.

max fu

71,124 次观看 • 3 个月前

Seniors discussing Architecture: Redis vs In-Memory Cache Meanwhile, the Vibe Coder😂 wait for it

Seniors discussing Architecture: Redis vs In-Memory Cache Meanwhile, the Vibe Coder😂 wait for it

Mashood K

152,890 次观看 • 4 个月前

RAM and CPU is DEAD ☠️

RAM and CPU is DEAD ☠️

kapilansh

928,280 次观看 • 2 个月前

Is it too much for you to handle?

Sensitive content

Is it too much for you to handle?

Fat Ass Sid

16,792 次观看 • 12 天前

RAM and CPU is DEAD ☠️😱

RAM and CPU is DEAD ☠️😱

Know Your Computer 💻

16,665 次观看 • 2 个月前

🚨 Today I published my own Stream Deck plugin on @Elgato Marketplace . I call it SimpleStats. It's a very lightweight plugin for viewing CPU usage, Ping, Net Throughput, GPU temps, even what is using up all your RAM!!!!!!! The most important part for me: Everything is included.

🚨 Today I published my own Stream Deck plugin on @Elgato Marketplace . I call it SimpleStats. It's a very lightweight plugin for viewing CPU usage, Ping, Net Throughput, GPU temps, even what is using up all your RAM!!!!!!! The most important part for me: Everything is included.

Brett

114,702 次观看 • 6 个月前

But the KV cache is created for each transformer layer. By sending each layer’s KV cache after it’s computed, we overlap communication with computation. We stream the KV cache and hide the network delay. We achieve a 4x speedup in prefill & 3x in decode, with 0 network delay.

But the KV cache is created for each transformer layer. By sending each layer’s KV cache after it’s computed, we overlap communication with computation. We stream the KV cache and hide the network delay. We achieve a 4x speedup in prefill & 3x in decode, with 0 network delay.

EXO Labs

50,617 次观看 • 1 年前

s1mple hitting awp clips on Cache what year is this 😍

s1mple hitting awp clips on Cache what year is this 😍

BC Game Esports

91,966 次观看 • 3 个月前

being able to pop into a level at an exact location is so important and so common, you NEED a button for it.

being able to pop into a level at an exact location is so important and so common, you NEED a button for it.

thomas "noio" vandenberg

20,938 次观看 • 7 个月前

Mommy asked mini Roshtein to do 69 with her but he is too small for it..

Mommy asked mini Roshtein to do 69 with her but he is too small for it..

Duel Moments

13,933 次观看 • 4 个月前