Brendan Gregg

NOVEMBER 29, 2017

It's all a bit confusing, and I wrote about this in 2014: [Xen Modes]. When I joined Netflix in 2014, we had begun transitioning from PV to HVM (PVHVM). But not all workloads: some are network bound (proxies) and storage bound (databases). ## 5. This recommendation quickly became out of date. The first was c3.

Virtualization 111

Virtualization AWS Hardware Storage 111

Uber Engineering

JANUARY 12, 2016

The making of Schemaless, Uber Engineering’s custom designed datastore using MySQL, which has allowed us to scale from 2014 to beyond. By Jakob Holdgaard Thomsen. This is part one of a three-part series on Schemaless.

Scalability 102

Scalability Design Engineering Database 102

SQL Server According to Bob

SEPTEMBER 26, 2016

As we moved towards SQL Server 2014, the pace of hardware accelerated. Our design for SQL Server 2012 and SQL Server 2014 is still proven and meets the demands for many of our customers. First, we looked at the overall architecture of the replica design. Direct seeding of new database replicas.

Availability 40

Availability Servers Hardware Benchmarking 40

SQL Performance

JULY 16, 2019

This made it easier for database professionals to make the case for a hardware upgrade, and made the typical upgrade more worthwhile. Intel officially moved on to what they call " Process- Architecture-Optimization (PAO) " in early 2016. You might be asking why you should care about all of this as a SQL Server Database professional?

Servers 46

Servers Hardware Architecture Retail 46

Uber Engineering

JANUARY 19, 2016

The details and examples of Schemaless triggers, a key feature of the datastore that’s kept Uber Engineering scaling since October 2014.

Engineering 94

Engineering Design Database Architecture 94

Alex Russell

MARCH 6, 2021

Devices and networks have evolved too: Alex Russell @slightlylate An update on mobile CPUs and the Performance Inequality Gap: Mid-tier Android devices (~$300) now get the single-core performance of a 2014 iPhone and the multi-core perf of a 2015 iPhone. mid-priced Androids were slightly faster than 2014's iPhone 6.

Performance 124

Performance Network Cache Metrics 124

Brendan Gregg

NOVEMBER 28, 2017

It's all a bit confusing, and I wrote about this in 2014: [Xen Modes]. When I joined Netflix in 2014, we had begun transitioning from PV to HVM (PVHVM). But not all workloads: some are network bound (proxies) and storage bound (databases). ## 5. This recommendation quickly became out of date. The first was c3.

Virtualization 40

Virtualization AWS Hardware Storage 40

SQL Performance

MARCH 8, 2021

SQL Server 2014 SP2 and later produce runtime (“actual”) execution plans that can include elapsed time and CPU usage for each execution plan operator (see KB3170113 and this blog post by Pedro Lopes). The following examples use the public Stack Overflow 2013 database ( download details ), with a single index added: CREATE INDEX PP ON dbo.

Servers 95

Servers Tuning Architecture Processing 95

The Netflix TechBlog

JANUARY 10, 2024

This architecture shift greatly reduced the processing latency and increased system resiliency. From Reloaded to Cosmos Reloaded Starting in 2014, we developed and operated the video processing pipeline on our third-generation platform Reloaded. The microservice architecture provides strong decoupling between services.

Processing 229

Processing Media Latency Innovation 229

Highly Scalable

MARCH 10, 2015

Simchi-Levi, 2014. RE94] Grouplens: an open architecture for collaborative filtering of netnews, P. VL02] The True Lift Model – A Novel Data Mining Approach to Response Modeling in Database Marketing, V. [GH02] Using Text Mining to Infer Semantic Attributes for Retail Data Mining, Rayid Ghani and Andrew E. Fano, 2002.

Retail 152

Retail C++ Analytics Metrics 152