Pinboard (jm)
https://pinboard.in/u:jm/public/
recent bookmarks from jmErasure Coding versus Tail Latency - Marc's Blog2023-05-23T22:05:23+00:00
https://brooker.co.za/blog/2023/01/06/erasure.html
jmarchitecture cache storage tail-latencies performance marc-brooker lambda erasure-coding algorithms latencyhttps://pinboard.in/https://pinboard.in/u:jm/b:dbbb79c2955e/Real-time analytics on network flow data with Apache Pinot2022-09-14T17:05:24+00:00
https://engineering.linkedin.com/blog/2022/real-time-analytics-on-network-flow-data-with-apache-pinot
jmInFlow requires storage of tens of TBs of data with a retention of 30 days. To support its real-time troubleshooting use case, the data must be queryable in real-time with sub-second latency so that engineers can query the data without any hassles during outages. For the storage layer, InFlow leverages Apache Pinot.
]]>pinot latency metrics linkedin network-flows realtime analytics storagehttps://pinboard.in/https://pinboard.in/u:jm/b:d408bc8a0e1e/Serverless Messaging: Latency Compared2022-09-07T08:46:07+00:00
https://bitesizedserverless.com/bite/serverless-messaging-latency-compared/
jmaws latency messaging queues architecture ops sqs sns kinesishttps://pinboard.in/https://pinboard.in/u:jm/b:0fd2c91e94ef/Numbers Every Programmer Should Know, By Year2020-03-03T14:41:26+00:00
https://colin-scott.github.io/personal_website/research/interactive_latency.html
jmmemory latency hardware history jeff-dean latencies speed performancehttps://pinboard.in/https://pinboard.in/u:jm/b:c7ea2fc97e3c/Circllhist2020-02-10T11:59:56+00:00
https://arxiv.org/pdf/2001.06561.pdf
jmhistograms aggregation quantiles percentiles measurement graphs data-structures summaries latency monitoring approximation papershttps://pinboard.in/https://pinboard.in/u:jm/b:763bb560c156/How to monitor Golden signals in Kubernetes2020-01-09T17:19:22+00:00
https://sysdig.com/blog/golden-signals-kubernetes/
jmkubernetes monitoring sysdig golden-data k8s golden-signals metrics latency errorshttps://pinboard.in/https://pinboard.in/u:jm/b:c8f356dfeae1/KIAM defaults result in massive latencies on AWS API calls2019-11-11T10:30:38+00:00
https://srvaroa.github.io/kubernetes/migration/latency/dns/java/aws/microservices/2019/10/22/kubernetes-added-a-0-to-my-latency.html
jmKIAM [a Kubernetes IAM API helper] happens to provide short-lived credentials to Pods, which makes sense as it’s fair to assume that the average lifetime of a Pod is shorter than EC2 instances. The default is precisely 15 min.
But if you put both defaults together, you have a problem. Each certificate provided to the application has a 15 min expiration time. The AWS Java SDK will force refreshing any certificate with less than 15 min expiration time left.
The result is that every request will be forced to refresh the temporary certificate, which requires two calls to the AWS API that add a huge latency penalty to each request. We later found a feature request in the AWS Java SDK that mentions this same issue.
The fix was easy. We reconfigured KIAM to request credentials with a longer expiration period. Once this change was applied, requests started being served without involving the AWS Metadata service and returned to an even lower latency than in EC2.
]]>kubernetes kiam defaults aws latency performance ec2https://pinboard.in/https://pinboard.in/u:jm/b:e6d5522f5c36/SQS performance and latency2017-07-10T15:03:12+00:00
https://softwaremill.com/amazon-sqs-performance-latency/?__s=gf36pf8g1gjugcqh6ppo
jmWe were looking at four values in the tests:
total number of messages sent per second (by all nodes)
total number of messages received per second
95th percentile of message send latency (how fast a message send call completes)
95th percentile of message processing latency (how long it takes between sending and receiving a message)
]]>sqs benchmarking measurement aws latencyhttps://pinboard.in/https://pinboard.in/u:jm/b:2438680a6eb8/AWS Inter-Region Latency Monitoring2017-06-12T09:51:31+00:00
https://www.cloudping.co/
jmlatency networking aws ops inter-region cross-region pinghttps://pinboard.in/https://pinboard.in/u:jm/b:ee529fc0773f/Amazon DynamoDB Accelerator (DAX)2017-04-20T10:36:09+00:00
https://aws.amazon.com/dynamodb/dax/
jmAmazon DynamoDB Accelerator (DAX) is a fully managed, highly available, in-memory cache for DynamoDB that delivers up to a 10x performance improvement – from milliseconds to microseconds – even at millions of requests per second. DAX does all the heavy lifting required to add in-memory acceleration to your DynamoDB tables, without requiring developers to manage cache invalidation, data population, or cluster management.
No latency percentile figures, unfortunately. Also still in preview.]]>amazon dynamodb aws dax performance storage databases latency low-latencyhttps://pinboard.in/https://pinboard.in/u:jm/b:cff8a162539c/AWS latency comparison: API Gateway vs Lambda vs Bare EC22016-10-12T11:04:22+00:00
https://www.prerender.cloud/lambda-latency
jmaws latency lambda api-gateway architecture httphttps://pinboard.in/https://pinboard.in/u:jm/b:afa22c5b2f10/ztellman/dirigiste2016-06-01T11:55:13+00:00
https://github.com/ztellman/dirigiste
jmasync jvm dirigiste java threadpools concurrency utilization capacity executors object-pools object-pooling latencyhttps://pinboard.in/https://pinboard.in/u:jm/b:5092c7f0c12e/Gil Tene on benchmarking2016-04-12T10:09:33+00:00
https://groups.google.com/forum/#!msg/mechanical-sympathy/0gaBXxFm4hE/O9QomwHIJAAJ
jmperformance benchmarking testing speed gil-tene latency measurement hdrhistogram load-testing loadhttps://pinboard.in/https://pinboard.in/u:jm/b:ddfcf3faf5bb/The Nyquist theorem and limitations of sampling profilers today, with glimpses of tracing tools from the future2016-02-09T12:55:23+00:00
http://danluu.com/perf-tracing/
jmThe cause [of some mystery widespread 250ms hangs] was kernel throttling of the CPU for processes that went beyond their usage quota. To enforce the quota, the kernel puts all of the relevant threads to sleep until the next multiple of a quarter second. When the quarter-second hand of the clock rolls around, it wakes up all the threads, and if those threads are still using too much CPU, the threads get put back to sleep for another quarter second. The phase change out of this mode happens when, by happenstance, there aren’t too many requests in a quarter second interval and the kernel stops throttling the threads. After finding the cause, an engineer found that this was happening on 25% of disk servers at Google, for an average of half an hour a day, with periods of high latency as long as 23 hours. This had been happening for three years. Dick Sites says that fixing this bug paid for his salary for a decade. This is another bug where traditional sampling profilers would have had a hard time. The key insight was that the slowdowns were correlated and machine wide, which isn’t something you can see in a profile.
]]>debugging performance visualization instrumentation metrics dan-luu latency google dick-sites linux scheduler throttling kernel hangshttps://pinboard.in/https://pinboard.in/u:jm/b:670193c955f5/League of Legends win-rates vs latency analysed2015-12-15T12:02:11+00:00
http://na.leagueoflegends.com/en/page/ping-winrate-and-vayne-probs
jmIt appears that more mechanically intensive champions are more affected by latency, while tankier champions or those with point-and-click abilities are less affected by latency.
(via Nelson)]]>games league-of-legends latency ping gaming internet via:nelsonhttps://pinboard.in/https://pinboard.in/u:jm/b:fc8a4454ba09/Low-latency journalling file write latency on Linux2015-12-09T14:58:39+00:00
http://epickrram.blogspot.co.uk/2015/12/journalling-revisited.html
jmlinux xfs ext3 ext4 filesystems lmax performance latency journalling opshttps://pinboard.in/https://pinboard.in/u:jm/b:9b225ebb15c4/ELS: latency based load balancer, part 12015-12-09T14:28:29+00:00
https://labs.spotify.com/2015/12/08/els-part-1/
jmELS measures the following things:
Success latency and success rate of each machine;
Number of outstanding requests between the load balancer and each machine. These are the requests that have been sent out but we haven’t yet received a reply;
Fast failures are better than slow failures, so we also measure failure latency for each machine.
Since users care a lot about latency, we prefer machines that are expected to answer quicker. ELS therefore converts all the measured metrics into expected latency from the client’s perspective.[...]
In short, the formula ensures that slower machines get less traffic and failing machines get much less traffic. Slower and failing machines still get some traffic, because we need to be able to detect when they come back up again.
]]>latency spotify proxies load-balancing els algorithms c3 round-robin load-balancers routinghttps://pinboard.in/https://pinboard.in/u:jm/b:067ff62fa178/Topics in High-Performance Messaging2015-12-02T15:53:00+00:00
https://www.informatica.com/downloads/1568_high_perf_messaging_wp/Topics-in-High-Performance-Messaging.htm
jmmessaging scalability scaling performance udp tcp protocols multicast latencyhttps://pinboard.in/https://pinboard.in/u:jm/b:aef1848d9376/Seastar2015-09-24T14:51:09+00:00
http://www.seastar-project.org/
jmc++ opensource performance framework scylla seastar latency linux shared-nothing multicorehttps://pinboard.in/https://pinboard.in/u:jm/b:44d66d8d03e7/toxy2015-08-28T11:19:51+00:00
https://github.com/h2non/toxy#latency
jmtoxy is a fully programmatic and hackable HTTP proxy to simulate server failure scenarios and unexpected network conditions. It was mainly designed for fuzzing/evil testing purposes, when toxy becomes particularly useful to cover fault tolerance and resiliency capabilities of a system, especially in service-oriented architectures, where toxy may act as intermediate proxy among services.
toxy allows you to plug in poisons, optionally filtered by rules, which essentially can intercept and alter the HTTP flow as you need, performing multiple evil actions in the middle of that process, such as limiting the bandwidth, delaying TCP packets, injecting network jitter latency or replying with a custom error or status code.
]]>toxy proxies proxy http mitm node.js soa network failures latency slowdown jitter bandwidth tcphttps://pinboard.in/https://pinboard.in/u:jm/b:4cc87b3a400a/Amazon EC2 2015 Benchmark: Testing Speeds Between AWS EC2 and S3 Regions2015-08-11T10:03:51+00:00
http://blog.takipi.com/amazon-ec2-2015-benchmark-testing-speeds-between-aws-ec2-and-s3-regions/
jmfail latency measurement aws ec2 percentiles s3https://pinboard.in/https://pinboard.in/u:jm/b:4e5264c55d58/Performance Testing at LMAX2015-06-03T12:49:58+00:00
http://epickrram.blogspot.co.uk/2014/05/performance-testing-at-lmax-part-one.html
jmperformance testing tests simulation latency lmax trading sniffing packet-capturehttps://pinboard.in/https://pinboard.in/u:jm/b:c962c123de77/SolarCapture Packet Capture Software2015-05-18T09:31:39+00:00
http://www.solarflare.com/SolarCapture-Packet-Capture-Software
jmSolarCapture is powerful packet capture product family that can transform every server into a precision network monitoring device, increasing network visibility, network instrumentation, and performance analysis. SolarCapture products optimize network monitoring and security, while eliminating the need for specialized appliances, expensive adapters relying on exotic protocols, proprietary hardware, and dedicated networking equipment.
See also Corvil (based in Dublin!): 'I'm using a Corvil at the moment and it's awesome- nanosecond precision latency measurements on the wire.'
(via mechanical sympathy list)]]>corvil timing metrics measurement latency network solarcapture packet-capture financial performance security network-monitoringhttps://pinboard.in/https://pinboard.in/u:jm/b:04ed77af0e50/"Trash Day: Coordinating Garbage Collection in Distributed Systems"2015-05-06T16:34:59+00:00
https://timharris.uk/papers/2015-hotos.pdf
jmblade via:adriancolyer papers gc distsys algorithms distributed java jvm latency spark cassandrahttps://pinboard.in/https://pinboard.in/u:jm/b:bf80614a2fb9/_Blade: a Data Center Garbage Collector_2015-04-17T09:46:59+00:00
http://arxiv.org/pdf/1504.02578.pdf
jmgc latency tail-latencies papers blade go java scheduling clustering load-balancing low-latency performancehttps://pinboard.in/https://pinboard.in/u:jm/b:604440c5b5eb/Gil Tene's "usual suspects" to reduce system-level hiccups/latency jitters in a Linux system2015-04-06T20:27:37+00:00
https://groups.google.com/d/msg/mechanical-sympathy/DWlziVmyW-w/at-54WECjL4J
jmBased on empirical evidence (across many tens of sites thus far) and note-comparing with others, I use a list of "usual suspects" that I blame whenever they are not set to my liking and system-level hiccups are detected. Getting these settings right from the start often saves a bunch of playing around (and no, there is no "priority" to this - you should set them all right before looking for more advice...).
]]>performance latency hiccups gil-tene tuning mechanical-sympathy hyperthreading linux opshttps://pinboard.in/https://pinboard.in/u:jm/b:e22d85fc3109/The Four Month Bug: JVM statistics cause garbage collection pauses (evanjones.ca)2015-03-26T22:16:21+00:00
http://www.evanjones.ca/jvm-mmap-pause.html
jmThe JVM by default exports statistics by mmap-ing a file in /tmp (hsperfdata). On Linux, modifying a mmap-ed file can block until disk I/O completes, which can be hundreds of milliseconds. Since the JVM modifies these statistics during garbage collection and safepoints, this causes pauses that are hundreds of milliseconds long. To reduce worst-case pause latencies, add the -XX:+PerfDisableSharedMem JVM flag to disable this feature. This will break tools that read this file, like jstat.
]]>bugs gc java jvm disk mmap latency ops jstathttps://pinboard.in/https://pinboard.in/u:jm/b:a8b7f0defa11/JClarity's Illuminate2015-02-26T15:49:43+00:00
http://www.jclarity.com/2015/02/24/why-we-built-illuminate-where-apm-is-going-next/
jmUsers download and install an Illuminate Daemon using a simple installer which starts up a small stand alone Java process. The Daemon sits quietly unless it is asked to start gathering SLA data and/or to trigger a diagnosis. Users can set SLA’s via the dashboard and can opt to collect latency measurements of their transactions manually (using our library) or by asking Illuminate to automatically instrument their code (Servlet and JDBC based transactions are currently supported).
SLA latency data for transactions is collected on a short cycle. When the moving average of latency measurements goes above the SLA value (e.g. 150ms), a diagnosis is triggered. The diagnosis is very quick, gathering key data from O/S, JVM(s), virtualisation and other areas of the system. The data is then run through the machine learned algorithm which will quickly narrow down the possible causes and gather a little extra data if needed.
Once Illuminate has determined the root cause of the performance problem, the diagnosis report is sent back to the dashboard and an alert is sent to the user. That alert contains a link to the result of the diagnosis which the user can share with colleagues. Illuminate has all sorts of backoff strategies to ensure that users don’t get too many alerts of the same type in rapid succession!
]]>illuminate jclarity java jvm scala latency gc tuning performancehttps://pinboard.in/https://pinboard.in/u:jm/b:d8f4dcc75538/Azul Zing on Ubuntu on AWS Marketplace2015-02-18T23:18:00+00:00
https://aws.amazon.com/marketplace/pp/B00M25ZRMC/ref=srh_res_product_title?ie=UTF8&sr=0-2&qid=1423925373136
jmzing azul latency performance ec2 awshttps://pinboard.in/https://pinboard.in/u:jm/b:093f5794163a/HdrHistogram: A better latency capture method2015-02-16T11:27:54+00:00
http://psy-lob-saw.blogspot.ie/2015/02/hdrhistogram-better-latency-capture.html
jmhdrhistogram hdr histograms statistics latency measurement metrics percentiles quantiles gil-tene nitsan-wakarthttps://pinboard.in/https://pinboard.in/u:jm/b:fe1e9f2ecc3d/Visualizing AWS Storage with Real-Time Latency Spectrogram2015-01-27T23:33:56+00:00
http://sysdigcloud.com/aws-storage-latency-sysdig-spectrogram/
jmdataviz latency io ops sysdig charts graphs commandline linuxhttps://pinboard.in/https://pinboard.in/u:jm/b:395dfb4e90a0/Comcast2015-01-27T16:15:05+00:00
https://github.com/tylertreat/Comcast
jmnetworking testing linux tc netem latency packet-loss iptableshttps://pinboard.in/https://pinboard.in/u:jm/b:99e8ef481d50/wrk22014-11-10T15:45:50+00:00
https://github.com/giltene/wrk2
jmwrk's model, which is similar to the model found in many current load generators, computes the latency for a given request as the time from the sending of the first byte of the request to the time the complete response was received. While this model correctly measures the actual completion time of individual requests, it exhibits a strong Coordinated Omission effect, through which most of the high latency artifacts exhibited by the measured server will be ignored. Since each connection will only begin to send a request after receiving a response, high latency responses result in the load generator coordinating with the server to avoid measurement during high latency periods.]]>wrk latency measurement tools cli http load-testing testing load-generation coordinated-omission gil-tenehttps://pinboard.in/https://pinboard.in/u:jm/b:2c02e86f370b/testing latency measurements using CTRL-Z2014-11-06T19:18:11+00:00
https://twitter.com/giltene/status/530432078639161344
jmGood example of why I always "calibrate" latency tools with ^Z tests. If ^Z results don't make sense, don't use [the] tool. ^Z test math examples: If you ^Z for half the time, Max is obvious. [90th percentile] should be 80% of the ^Z stall time.
]]>control-z suspend unix testing latencies latency measurement percentiles tipshttps://pinboard.in/https://pinboard.in/u:jm/b:588d9f8b5c90/Linux kernel's Transparent Huge Pages feature causing 300ms-800ms pauses2014-11-04T18:05:19+00:00
https://groups.google.com/a/jclarity.com/forum/#!topic/friends/QPXl_8ce5Fg
jmredis memory defrag huge-pages linux kernel ops latency performance transparent-huge-pageshttps://pinboard.in/https://pinboard.in/u:jm/b:02aad4ff8ea8/Testing fork time on AWS/Xen infrastructure2014-10-30T16:03:54+00:00
https://redislabs.com/blog/testing-fork-time-on-awsxen-infrastructure#.VFJgrFOsXFd
jmfork xen redis bugs performance latency p99https://pinboard.in/https://pinboard.in/u:jm/b:097ee6064211/Most page loads will experience the 99th percentile response latency2014-10-29T12:25:45+00:00
http://latencytipoftheday.blogspot.ie/2014/06/latencytipoftheday-most-page-loads.html?m=1
jmMOST of the page view attempts will experience the 99%'lie server response time in modern web applications. You didn't read that wrong.
]]>latency metrics percentiles p99 web http soahttps://pinboard.in/https://pinboard.in/u:jm/b:46b1477c7ddd/FelixGV/tehuti2014-10-09T10:53:00+00:00
https://github.com/FelixGV/tehuti
jmasl2 apache open-source tehuti metrics percentiles quantiles statistics measurement latency kafka voldemort linkedinhttps://pinboard.in/https://pinboard.in/u:jm/b:a2f55ebce7bb/"Quantiles on Streams" [paper, 2009]2014-10-08T09:49:58+00:00
http://www.cs.ucsb.edu/~suri/psdir/ency.pdf
jmlatency percentiles coding quantiles streams papers algorithmshttps://pinboard.in/https://pinboard.in/u:jm/b:662952253d16/Tehuti2014-10-08T09:45:50+00:00
https://groups.google.com/forum/#!msg/project-voldemort/Y52UyHQ8tBA/9Ei79_RvS3EJ
jmkafka metrics dropwizard java scala jvm timers ewma statistics measurement latency sampling tehuti voldemort linkedin jay-krepshttps://pinboard.in/https://pinboard.in/u:jm/b:b56664c1a098/"Left-Right: A Concurrency Control Technique with Wait-Free Population Oblivious Reads" [pdf]2014-09-17T14:29:55+00:00
http://sourceforge.net/projects/ccfreaks/files/papers/LeftRight/leftright-extended.pdf
jmleft-right concurrency multithreading wait-free blocking realtime gc latency reader-writer locking synchronization javahttps://pinboard.in/https://pinboard.in/u:jm/b:71e20ba2b8b1/"The Tail at Scale"2014-07-08T16:58:49+00:00
http://prophet.cs.duke.edu/courses/compsci590.4/fall13/838-CloudPapers/dean_longtail.pdf
jmgoogle architecture distcomp soa http partitioning replication latency 99th-percentile canary-requests hedged-requestshttps://pinboard.in/https://pinboard.in/u:jm/b:eb38667ad2a8/How to perform a load/latency test, correcting for coordinated-omission error2014-07-03T14:05:19+00:00
https://groups.google.com/forum/#!msg/mechanical-sympathy/g-iCw1HbZ-o/XXcQYb596X4J
jmgil-tene coordinated-omission measurement jmh latency testing errors codehttps://pinboard.in/https://pinboard.in/u:jm/b:c98b6e03164d/Google's Pegasus2014-06-18T15:48:28+00:00
http://csl.stanford.edu/~christos/publications/2014.pegasus.isca.pdf
jmpegasus power-management power via:fanf google latency scalinghttps://pinboard.in/https://pinboard.in/u:jm/b:ddece9bbccdf/Monitoring Reactive Applications with Kamon2014-05-19T10:41:38+00:00
http://kamon.io/presentations/javacro14/#/
jmmetrics dropwizard hdrhistogram gil-tene kamon akka spray play reactive statistics java scala percentiles latencyhttps://pinboard.in/https://pinboard.in/u:jm/b:dd1798009833/Uplink Latency of WiFi and 4G Networks2014-04-22T22:33:14+00:00
http://www.igvita.com/2014/04/21/uplink-scheduling-in-4G-networks/
jmwifi 3g 4g mobile networking internet latency tcphttps://pinboard.in/https://pinboard.in/u:jm/b:2add3215a9fb/Game servers: UDP vs TCP2014-04-01T21:49:21+00:00
https://news.ycombinator.com/item?id=7507377
jmudp tcp games protocols networking latency internet gaming hackernewshttps://pinboard.in/https://pinboard.in/u:jm/b:1876270a5c09/Micro jitter, busy waiting and binding CPUs2014-03-21T11:26:50+00:00
http://vanillajava.blogspot.co.uk/2013/07/micro-jitter-busy-waiting-and-binding.html?m=1
jmpinning threads performance latency jitter tuninghttps://pinboard.in/https://pinboard.in/u:jm/b:186b39854e95/'Bobtail: Avoiding Long Tails in the Cloud' [pdf]2014-02-27T15:15:06+00:00
https://www.usenix.org/system/files/conference/nsdi13/nsdi13-final77.pdf
jmnetworking ec2 bobtail latency long-tail xen performancehttps://pinboard.in/https://pinboard.in/u:jm/b:a724de5104f3/LatencyUtils by giltene2013-11-13T22:00:09+00:00
http://latencyutils.github.io/LatencyUtils/
jmThe LatencyUtils package includes useful utilities for tracking latencies. Especially in common in-process recording scenarios, which can exhibit significant coordinated omission sensitivity without proper handling.
]]>gil-tene metrics java measurement coordinated-omission latency speed service-metrics open-sourcehttps://pinboard.in/https://pinboard.in/u:jm/b:cee7964c8c45/"Effective Computation of Biased Quantiles over Data Streams" [paper]2013-11-01T16:57:06+00:00
http://www.cs.rutgers.edu/~muthu/bquant.pdf
jm
Skew is prevalent in many data sources such as IP traffic streams.To continually summarize the distribution of such data, a high-biased set of quantiles (e.g., 50th, 90th and 99th percentiles) with finer error guarantees at higher ranks (e.g., errors of 5, 1 and 0.1 percent, respectively) is more useful than uniformly distributed quantiles (e.g., 25th, 50th and 75th percentiles) with uniform error guarantees. In this paper, we address the following two prob-lems. First, can we compute quantiles with finer error guarantees for the higher ranks of the data distribution effectively, using less space and computation time than computing all quantiles uniformly at the finest error? Second, if specific quantiles and their error bounds are requested a priori, can the necessary space usage and computation time be reduced? We answer both questions in the affirmative by formalizing them as the “high-biased” quantiles and the “targeted” quantiles problems, respectively, and presenting algorithms with provable guarantees, that perform significantly better than previously known solutions for these problems. We implemented our algorithms in the Gigascope data stream management system, and evaluated alternate approaches for maintaining the relevant summary structures.Our experimental results on real and synthetic IP data streams complement our theoretical analyses, and highlight the importance of lightweight, non-blocking implementations when maintaining summary structures over high-speed data streams.
Implemented as a timer-histogram storage system in http://armon.github.io/statsite/ .]]>
statistics quantiles percentiles stream-processing skew papers histograms latency algorithmshttps://pinboard.in/https://pinboard.in/u:jm/b:54346b7d58f0/Barbarians at the Gateways - ACM Queue2013-10-19T20:36:17+00:00
http://queue.acm.org/detail.cfm?id=2536492
jm
I am a former high-frequency trader. For a few wonderful years I led a group of brilliant engineers and mathematicians, and together we traded in the electronic marketplaces and pushed systems to the edge of their capability.
Insane stuff -- FPGAs embedded in the network switches to shave off nanoseconds of latency.]]>low-latency hft via:nelson markets stock-trading latency fpgas networkinghttps://pinboard.in/https://pinboard.in/u:jm/b:89014695d9aa/"High Performance Browser Networking", by Ilya Grigorik, read online for free2013-10-15T09:57:32+00:00
http://chimera.labs.oreilly.com/books/1230000000545
jmHow prepared are you to build fast and efficient web applications? This eloquent book provides what every web developer should know about the network, from fundamental limitations that affect performance to major innovations for building even more powerful browser applications—including HTTP 2.0 and XHR improvements, Server-Sent Events (SSE), WebSocket, and WebRTC.
Author Ilya Grigorik, a web performance engineer at Google, demonstrates performance optimization best practices for TCP, UDP, and TLS protocols, and explains unique wireless and mobile network optimization requirements. You’ll then dive into performance characteristics of technologies such as HTTP 2.0, client-side network scripting with XHR, real-time streaming with SSE and WebSocket, and P2P communication with WebRTC.
Deliver optimal TCP, UDP, and TLS performance;
Optimize network performance over 3G/4G mobile networks;
Develop fast and energy-efficient mobile applications;
Address bottlenecks in HTTP 1.x and other browser protocols;
Plan for and deliver the best HTTP 2.0 performance;
Enable efficient real-time streaming in the browser;
Create efficient peer-to-peer videoconferencing and low-latency applications with real-time WebRTC transports
Via Eoin Brazil.]]>book browser networking performance phones mobile 3g 4g hsdpa http udp tls ssl latency webrtc websockets ebooks via:eoin-brazil google http2 sse xhr ilya-grigorikhttps://pinboard.in/https://pinboard.in/u:jm/b:5312bd9b166a/Rapid read protection in Cassandra 2.0.22013-10-05T20:56:35+00:00
http://www.datastax.com/dev/blog/rapid-read-protection-in-cassandra-2-0-2
jmcassandra nosql replication distcomp latency storagehttps://pinboard.in/https://pinboard.in/u:jm/b:74b80ef24320/Attacking Tor: how the NSA targets users' online anonymity2013-10-04T16:31:54+00:00
http://www.theguardian.com/world/2013/oct/04/tor-attacks-nsa-users-online-anonymity
jmAs part of the Turmoil system, the NSA places secret servers, codenamed Quantum, at key places on the internet backbone. This placement ensures that they can react faster than other websites can. By exploiting that speed difference, these servers can impersonate a visited website to the target before the legitimate website can respond, thereby tricking the target's browser to visit a Foxacid server.
whoa, I missed this before.]]>nsa gchq packet-injection attacks security backbone http latencyhttps://pinboard.in/https://pinboard.in/u:jm/b:0a1aea658272/Why Tellybug moved from Cassandra to Amazon DynamoDB2013-10-02T12:55:23+00:00
http://attentionshard.wordpress.com/2013/09/30/why-tellybug-moved-from-cassandra-to-amazon-dynamodb/
jmaws dynamodb cassandra nosql storage tellybug counters scalability reliability latencyhttps://pinboard.in/https://pinboard.in/u:jm/b:8b0a38474b92/Benchmarking Redis on AWS ElastiCache2013-09-16T22:48:55+00:00
http://blog.meldium.com/home/2013/9/13/benchmarking-redis-on-aws-elasticache
jmlatency redis measurement benchmarks ec2 elasticache aws storage testshttps://pinboard.in/https://pinboard.in/u:jm/b:19ca978874ad/[#CASSANDRA-5582] Replace CustomHsHaServer with better optimized solution based on LMAX Disruptor2013-09-03T22:29:45+00:00
https://issues.apache.org/jira/browse/CASSANDRA-5582
jmdisruptor cassandra java p99 latency speed performance concurrency via:kellabytehttps://pinboard.in/https://pinboard.in/u:jm/b:70d101c721e7/Coordinated Omission2013-08-05T22:00:16+00:00
https://groups.google.com/forum/#!msg/mechanical-sympathy/icNZJejUHfE/BfDekfBEs_sJ
jm
I've been harping for a while now about a common measurement technique problem I call "Coordinated Omission" for a while, which can often render percentile data useless. [...] I believe that this problem occurs extremely frequently in test results, but it's usually hard to deduce it's existence purely from the final data reported. But every once in a while, I see test results where the data provided is enough to demonstrate the huge percentile-misreporting effect of Coordinated Omission based purely on the summary report.
I ran into just such a case in Attila's cool posting about log4j2's truly amazing performance, so I decided to avoid polluting his thread with an elongated discussion of how to compute 99.9%'ile data, and started this topic here. That thread should really be about how cool log4j2 is, and I'm certain that it really is cool, even after you correct the measurements. [...] Basically, I think that the 99.99% observation computation is wrong, and demonstrably (using the data in the graph data posted) exhibits the classic "coordinated omission" measurement problem I've been preaching about. This test is not alone in exhibiting this, and there is nothing to be ashamed of when you find yourself making this mistake. I only figured it out after doing it myself many many times, and then I noticed that everyone else seems to also be doing it but most of them haven't yet figured it out. In fact, I run into this issue so often in percentile reporting and load testing that I'm starting to wonder if coordinated omission is there in 99.9% of latency tests ;-)
]]>measurement testing latency load-testing gil-tene coordinated-omission validity log4j percentileshttps://pinboard.in/https://pinboard.in/u:jm/b:36240fa70ac9/Improved HTTPS Performance with Early SSL Termination2013-07-15T09:50:00+00:00
http://blog.filepicker.io/post/29422604907/improved-https-performance-with-early-ssl-termination
jmhttp https ssl architecture aws ec2 performance latency internet round-trip nginx tlshttps://pinboard.in/https://pinboard.in/u:jm/b:c5687e3aabe7/_Measuring Mobile Web Performance_ [slides]2013-06-28T11:27:12+00:00
http://www.slideshare.net/sthair/measuring-mobile-web-performance-v2
jmrtt packets latency hsdpa mobile internet trains packet-losshttps://pinboard.in/https://pinboard.in/u:jm/b:a63232d61e87/SSL/TLS overhead2013-06-21T14:43:55+00:00
http://netsekure.org/2010/03/tls-overhead/
jmnetwork tls ssl performance latency speed networking internet security packets tcp handshakehttps://pinboard.in/https://pinboard.in/u:jm/b:9941807de694/Communication costs in real-world networks2013-05-24T16:08:04+00:00
http://www.bailis.org/blog/communication-costs-in-real-world-networks/
jm
I wasn’t aware of any datasets describing network behavior both within and across datacenters, so we launched m1.small Amazon EC2 instances in each of the eight geo-distributed “Regions,” across the three us-east “Availability Zones” (three co-located datacenters in Virginia), and within one datacenter (us-east-b). We measured RTTs between hosts for a week at a granularity of one ping per second.
Some of the high-percentile measurements are undoubtedly impact of host and VM behaviour, but that is still good data for a typical service built in EC2.]]>networks performance measurements benchmarks ops ec2 networking internet az latencyhttps://pinboard.in/https://pinboard.in/u:jm/b:9dade9cdeb00/Storm and Hadoop: Convergence of Big-Data and Low-Latency Processing2013-02-28T09:57:37+00:00
http://developer.yahoo.com/blogs/ydn/posts/2013/02/storm-and-hadoop-convergence-of-big-data-and-low-latency-processing/
jmyahoo yarn cloud-computing private-clouds big-data latency storm hadoop elastic-computing hbasehttps://pinboard.in/https://pinboard.in/u:jm/b:350773902d3f/#AltDevBlogADay » Latency Mitigation Strategies2013-02-22T17:55:55+00:00
http://www.altdevblogaday.com/2013/02/22/latency-mitigation-strategies/
jm
Virtual reality (VR) is one of the most demanding human-in-the-loop applications from a latency standpoint. The latency between the physical movement of a user’s head and updated photons from a head mounted display reaching their eyes is one of the most critical factors in providing a high quality experience.
Human sensory systems can detect very small relative delays in parts of the visual or, especially, audio fields, but when absolute delays are below approximately 20 milliseconds they are generally imperceptible. Interactive 3D systems today typically have latencies that are several times that figure, but alternate configurations of the same hardware components can allow that target to be reached.
A discussion of the sources of latency throughout a system follows, along with techniques for reducing the latency in the processing done on the host system.
]]>head-mounted-display display ui latency vision coding john-carmackhttps://pinboard.in/https://pinboard.in/u:jm/b:8f9b7e07fe99/AnandTech - The Intel SSD DC S3700: Intel's 3rd Generation Controller Analyzed2012-11-06T11:01:06+00:00
http://www.anandtech.com/show/6432/the-intel-ssd-dc-s3700-intels-3rd-generation-controller-analyzed/3
jmlatency intel via:martin-thompson optimization speed p99 data-structures arrays btrees ssd hardwarehttps://pinboard.in/https://pinboard.in/u:jm/b:04d3d7d3e082/How does LMAX's disruptor pattern work? - Stack Overflow2011-11-23T09:52:35+00:00
http://stackoverflow.com/questions/6559308/how-does-lmaxs-disruptor-pattern-work
jmvia:filippo servers seda queueing concurrency disruptor patterns latency trading performance ring-buffershttps://pinboard.in/https://pinboard.in/u:jm/b:388ee9929e15/Overclocking SSL2010-07-05T12:50:29+00:00
http://www.imperialviolet.org/2010/06/25/overclocking-ssl.html
jmcertificates encryption google https latency speed ssl tcp tls web performancehttps://pinboard.in/u:jm/b:52be96613038/MapReduce as a way to cope with high-latency memory2010-06-21T11:21:46+00:00
http://lists.canonical.org/pipermail/kragen-tol/2010-June/000917.html
jmkragen thoughts random mapreduce memory speed latencyhttps://pinboard.in/u:jm/b:e82fa500b315/