On 11/11/06, Ryan Barrett <[EMAIL PROTECTED]> wrote:
... i'm glad you brought this up. it's a huge point, and it's not limited to disks. latency has lagged bandwidth by orders of magnitude in memory, network connections, and CPU cores for decades
one of my favorite examples is SIFS/DIFS timing in 802.11 for long point-to-point links[0]. just about everything "high performance" these days is spending considerable effort to shave latency tolerances to absolute minimums. i suppose this is really an old problem reinvented regularly (think connection machines and hyper cube geometries...)
there are standard ways to mitigate this, including caching, prefetching, and prediction. we could definitely use them more aggressively. in many latency sensitive areas at my work, we've given up on disks altogether and serve purely out of memory. that's clearly only a short term fix, though. i've been itching to work on a project in this area for a long time, but haven't thought of an idea worth pursuing yet. any ideas?
i rarely pass up an opportunity to beat the implicit feedback horse (corpse?). an early favorite is the elephant file system[1]. the question they pose, "when to forget?" is answered by the same techniques that can direct caching and prefetching. take these implicit methods[2] and apply to your favorite problem, whether ordering peers by reputation for use in iterative unicast discovery, or background cache filling of resources ordered by individual relevance metrics. sometimes explicit weights / heuristics can improve usability or performance, but i find multi-faceted implicit methods far more robust and resilient; well worth the design and implementation complexity required. 0. "Long Host WiFi Timing" http://web.archive.org/web/20050206183432/peertech.org/LongShotWiFiTiming 1. "Deciding when to forget in the Elephant file system (1999)" http://citeseer.ist.psu.edu/santry99deciding.html 2. "Implicit Feedback" http://zgp.org/pipermail/p2p-hackers/2006-April/003786.html best regards, _______________________________________________ p2p-hackers mailing list [email protected] http://lists.zooko.com/mailman/listinfo/p2p-hackers
