article thumbnail

The Surprising Effectiveness of Non-Overlapping, Sensitivity-Based Performance Models

John McCalpin

This data is from the 2007 presentation. It is not surprising that there is a lot of scatter, but the factor of four range in Peak MFLOPS at fixed SPECfp_rate2000/core and the factor of four range in SPECfp_rate2000/core at fixed Peak MFLOPS was higher than I expected… (Also from the 2007 presentation.)

article thumbnail

The Best In Performance Interview Series – Episode #4: Recap with Rich Howard

Rigor

When working on a project for the Man Booker Prize in 2007 (now known as the Booker Prize), Howard saw first-hand how a heavy-load day could impact a website. Make technical changes like increasing the length of cache headers within their particular assets. Retrospectives. Lessons learned. From there, he will calculate a load model.

article thumbnail

Egnyte Architecture: Lessons learned in building and scaling a multi petabyte content platform

High Scalability

Egnyte is a secure Content Collaboration and Data Governance platform, founded in 2007 when Google drive wasn't born and AWS S3 was cost-prohibitive. Edge caching. In general, Egnyte connect architecture shards and caches data at different levels based on: Amount of data. Nginx for disk based caching. Hybrid Sync.