The Morning Paper

JULY 4, 2019

Are you ready to take your system assurance programme to the next level? In all cases we need to be able to carefully monitor the impact on the system, and back out if things start going badly wrong. Netflix’s system is deployed on the public cloud as complex set of interacting microservices.

Latency 77

Latency Engineering Metrics Traffic 77

Brendan Gregg

JUNE 29, 2018

sounds like a homework exercise of purely academic value. I've refuted many benchmarks by showing that they would require a network throughput that would far exceed the maximum network bandwidth (off by, for example, as much as 10x!). Networking is the easiest to check. This is really asking "what's the limiter?"

Benchmarking 40

Benchmarking Latency Cache Network 40

The Morning Paper

NOVEMBER 14, 2019

It’s another networking paper to close out the week (and our coverage of SOSP’19), but whereas Snap looked at traffic routing within the datacenter, Taiji is concerned with routing traffic from the edge to a datacenter. For example, balance utilisation across all data centers, or optimise for network latency. SOSP’19.

Traffic 42

Traffic Internet Internet Latency 42

All Things Distributed

OCTOBER 2, 2017

With these requirements in mind, and a willingness to question the status quo, a small group of distributed systems experts came together and designed a horizontally scalable distributed database that would scale out for both reads and writes to meet the long-term needs of our business. This was the genesis of the Amazon Dynamo database.

Internet 128

Internet Internet AWS Performance 128

Adrian Cockcroft

NOVEMBER 11, 2019

A resilient system continues to operate successfully in the presence of failures. There are many possible failure modes, and each exercises a different aspect of resilience. Hence, one way to reduce risk is to make systems more observable. This discussion focuses on hardware, software and operational failure modes.

Latency 52

Latency Engineering Systems Hardware 52

John McCalpin

AUGUST 1, 2018

This is an intellectually challenging and labor-intensive exercise, requiring detailed review of the published details of each of the components of the system, and usually requiring significant “detective work” (using customized microbenchmarks, hardware performance counter analysis, and creative thinking) to fill in the gaps.

Hardware 40

Hardware Transportation Performance Latency 40

Adrian Cockcroft

NOVEMBER 11, 2019

A resilient system continues to operate successfully in the presence of failures. There are many possible failure modes, and each exercises a different aspect of resilience. Hence, one way to reduce risk is to make systems more observable. This discussion focuses on hardware, software and operational failure modes.

Latency 53

Latency Engineering Systems Hardware 53

The Morning Paper

DECEMBER 8, 2019

At the start of November I was privileged to attend HPTS (the High Performance Transaction Systems) conference in Asilomar. Byte-addressable non-volatile memory,) NVM will fundamentally change the way hardware interacts, the way operating systems are designed, and the way applications operate on data. PLOS’19.

Hardware 56

Hardware Programming Media Storage 56

All Things Distributed

NOVEMBER 15, 2010

Werner Vogels weblog on building scalable and robust distributed systems. For example, the most fundamental abstraction trade-off has always been latency versus throughput. The throughput of this pipeline is more important than the latency of the individual operations. All Things Distributed. Comments ().

AWS 136

AWS Latency Programming Architecture 136