TE
TechEcho
Home24h TopNewestBestAskShowJobs
GitHubTwitter
Home

TechEcho

A tech news platform built with Next.js, providing global tech news and discussions.

GitHubTwitter

Home

HomeNewestBestAskShowJobs

Resources

HackerNews APIOriginal HackerNewsNext.js

© 2025 TechEcho. All rights reserved.

New cache design speeds up processing time by 15%

94 pointsby srikarabout 11 years ago

4 comments

AceJohnny2about 11 years ago
Reminds me of the hilarious column by James Mickens, &quot;The Slow Winter&quot;, on how these kinds of improvements used to be easy:<p>&quot;“I wish that we could predict branches more accurately,” and you’d think, “maybe we can leverage three bits of state per branch to implement a simple saturating counter,” and you’d laugh and declare that such a stupid scheme would never work, but then you’d test it and it would be 94% accurate, and the branches would wake up the next morning and read their newspapers and the headlines would say OUR WORLD HAS BEEN SET ON FIRE.&quot;<p>PDF: <a href="https://www.usenix.org/system/files/1309_14-17_mickens.pdf" rel="nofollow">https:&#x2F;&#x2F;www.usenix.org&#x2F;system&#x2F;files&#x2F;1309_14-17_mickens.pdf</a>
评论 #7432036 未加载
habosaabout 11 years ago
I just recently took an advanced computer architecture course, and I was stunned by how little I knew about how computers ~actually~ work.<p>Most people think &quot;memory is fast, disk is slow&quot; but the reality is &quot;high-level caches are fast, other caches are slow, memory is really slow, and stay the hell away from disk&quot;. You&#x27;re always taught to think in terms of memory but the good news is that very intelligent cache designers have made accessing memory directly a relatively infrequent event, and that&#x27;s one of the only reasons our processors get to do any work at all.
评论 #7431053 未加载
评论 #7430149 未加载
评论 #7431674 未加载
solarexplorerabout 11 years ago
Link to the paper: <a href="http://people.csail.mit.edu/devadas/pubs/acc-hpca14.pdf" rel="nofollow">http:&#x2F;&#x2F;people.csail.mit.edu&#x2F;devadas&#x2F;pubs&#x2F;acc-hpca14.pdf</a>
valarauca1about 11 years ago
Cache misses are a major problem many people don&#x27;t think about or shy away from because you just need to assume they&#x27;ll happen.<p>Missing L1 cache costs between 10-60 cycles, your 10x that for L2. Access to the direct RAM because you also missed L3? you start dealing in 100,000+ processor cycles of hang.<p>Glad to see were moving forward.
评论 #7429620 未加载
评论 #7430458 未加载
评论 #7429878 未加载
评论 #7429496 未加载