All Things Distributed

DECEMBER 18, 2012

The Back-to-Basics Readings of 2012. By Werner Vogels on 18 December 2012 10:00 PM. May 28 - Virtual Time and Global States of Distributed Systems , Friedemann Mattern, Parallel and Distributed Algorithms, North-Holland (1989) , p. All Things Distributed. Werner Vogels weblog on building scalable and robust distributed systems.

Operating System 144

Operating System Cache AWS Scalability 144

Brendan Gregg

FEBRUARY 28, 2023

Titus, the Netflix container management platform, is now open source,” [link] Apr 2018 - [Cutress 19] Dr. DDR6: Here's What to Expect in RAM Modules,” [link] Nov 2020 - [Salter 20] Jim Salter, “Western Digital releases new 18TB, 20TB EAMR drives,” [link] Jul 2020 - [Spier 20] Martin Spier, Brendan Gregg, et al.,

Performance 110

Performance Latency Cache Virtualization 110

Brendan Gregg

FEBRUARY 28, 2023

Titus, the Netflix container management platform, is now open source,” [link] , Apr 2018 [Cutress 19] Dr. DDR6: Here's What to Expect in RAM Modules,” [link] , Nov 2020 [Salter 20] Jim Salter, “Western Digital releases new 18TB, 20TB EAMR drives,” [link] , Jul 2020 [Spier 20] Martin Spier, Brendan Gregg, et al.,

Performance 52

Performance Latency Cache Virtualization 52

ScaleOut Software

MAY 15, 2014

For example, the IMDG must be able to efficiently create millions of objects in each server to make use of its huge storage capacity. Given all this, we thought it would be a good opportunity to see how we are doing relative to the competition, and in particular, relative to Microsoft’s AppFabric caching for Windows on-premise servers.

Cache 40

Cache Servers Network Design 40

Smashing Magazine

JUNE 30, 2021

In terms for performance, these phones have plateaued at the performance of a 2012 iPhone. Green Energy Efficient Progressive Web Apps ,” David, Microsoft DevBlogs. As a rule of thumb: If JSON.stringify(messagePayload) is under 10KB, you are at virtually no risk of creating a long frame, even on the slowest of phones.

Games 145

Games Architecture Code C++ 145

All Things Distributed

JANUARY 18, 2012

By Werner Vogels on 18 January 2012 07:00 AM. This is not just predictability of median performance and latency, but also at the end of the distribution (the 99.9th percentile), so we could provide acceptable performance for virtually every customer. Additional request capacity is priced at cost-efficiently hourly rates as low as $.01

Scalability 167

Scalability Database Ecommerce Latency 167