{"id":647,"date":"2011-04-21T19:45:13","date_gmt":"2011-04-21T17:45:13","guid":{"rendered":"http:\/\/cerezo.name\/blog\/?p=647"},"modified":"2024-10-14T14:19:44","modified_gmt":"2024-10-14T12:19:44","slug":"lmax-going-slow-to-go-fast","status":"publish","type":"post","link":"http:\/\/cerezo.name\/blog\/2011\/04\/21\/lmax-going-slow-to-go-fast\/","title":{"rendered":"LMAX<\/span>: Going Slow to Go Fast"},"content":{"rendered":"\n
GDE<\/span> Error: Error retrieving file \u2014 if necessary turn off error checking (404:Not Found)<\/div>\n\n

Scalable web architects must learn from the transactional world, stock exchanges and other financial creatures of mass data processing. For example, LMAX<\/span> attaining 100K<\/span>+ TPS<\/span> at less than 1 ms latency is a remarkable technical feat. Sure, other exchanges like NYSE<\/span> and LSE<\/span> manage to achieve higher TPS<\/span> at lower latencies, however it\u2019s the history behind LMAX<\/span> what makes it a fascinating object of study: I\u2019ve estimated from public sources that 20 people worked fully committed for three years to develop the initially released version of LMAX<\/span>. At first, a simple proof of concept reaching 10K<\/span> TPS<\/span> was produced, followed by a long number recode-measure-debug cycles that felt more like squeezing and juicing the JVM<\/span> to achieve significant speed-ups than real programming, because writing cache-friendly code for an adaptive-optimizing JIT<\/span> virtual machine with no control of how data structures are mapped to memory is really hard, as nonlinearities appear everywhere in the typical code optimization process.<\/p>\n

It\u2019s the same kind of technical debt Google experienced<\/a> when it started with a Java codebase, then migrated to a slower Python one to finally settle for C\/C++; or the current technical debt at Twitter, a pure Ruby on Rails product that moved to Java <\/a> and Scala<\/a> with phenomenal results.<\/p>\n

When frameworks and virtual machines get in the way, it\u2019s the good old wisdom from people like Ken Thompson that illuminates the path to success: \u201cOne of my most productive days was throwing away 1000 lines of code.\u201d<\/p>\n","protected":false},"excerpt":{"rendered":"

Scalable web architects must learn from the transactional world, stock exchanges and other financial creatures of mass data processing. For example, LMAX<\/span> attaining 100K<\/span>+ TPS<\/span> at less than 1 ms latency is a remarkable technical feat. Sure, other exchanges like NYSE<\/span> and LSE<\/span> manage to achieve higher TPS<\/span> at lower latencies, however it\u2019s the history [\u2026]<\/p>\n","protected":false},"author":1,"featured_media":0,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"wp_typography_post_enhancements_disabled":false,"ngg_post_thumbnail":0},"categories":[14,15],"tags":[],"_links":{"self":[{"href":"http:\/\/cerezo.name\/blog\/wp-json\/wp\/v2\/posts\/647"}],"collection":[{"href":"http:\/\/cerezo.name\/blog\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"http:\/\/cerezo.name\/blog\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"http:\/\/cerezo.name\/blog\/wp-json\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"http:\/\/cerezo.name\/blog\/wp-json\/wp\/v2\/comments?post=647"}],"version-history":[{"count":5,"href":"http:\/\/cerezo.name\/blog\/wp-json\/wp\/v2\/posts\/647\/revisions"}],"predecessor-version":[{"id":1646,"href":"http:\/\/cerezo.name\/blog\/wp-json\/wp\/v2\/posts\/647\/revisions\/1646"}],"wp:attachment":[{"href":"http:\/\/cerezo.name\/blog\/wp-json\/wp\/v2\/media?parent=647"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"http:\/\/cerezo.name\/blog\/wp-json\/wp\/v2\/categories?post=647"},{"taxonomy":"post_tag","embeddable":true,"href":"http:\/\/cerezo.name\/blog\/wp-json\/wp\/v2\/tags?post=647"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}