Brendan Gregg

FEBRUARY 28, 2023

Make sure your system can handle next-generation DRAM,” [link] , Nov 2011 [Hruska 12] Joel Hruska, “The future of CPU scaling: Exploring options on the cutting edge,” [link] , Feb 2012 [Gregg 13] Brendan Gregg, “Blazing Performance with Flame Graphs,” [link] , 2013 [Shimpi 13] Anand Lal Shimpi, “Seagate to Ship 5TB HDD in 2014 using Shingled Magnetic (..)

Performance 52

Performance Latency Cache Virtualization 52

All Things Distributed

JULY 19, 2012

By Werner Vogels on 19 July 2012 12:00 AM. With PV virtualization we are seeing between 10,000 and 85,000 IOPS for 4k random writes and with HVM between 9,000 and 75,000. Increasingly randomized access and burst IO through aggregation put strains on any IO subsystem, physical or virtual, attached or remote. Comments ().

Cloud 129

Cloud Performance AWS Virtualization 129

All Things Distributed

JANUARY 23, 2012

By Werner Vogels on 23 January 2012 12:01 AM. The Amazon Virtual Private Cloud extends on-premises compute with all the power of AWS, making it elastic, scalable and highly reliable. All Things Distributed. Werner Vogels weblog on building scalable and robust distributed systems. Expanding the Cloud - The AWS Storage Gateway.

Storage 115

Storage AWS Virtualization Cloud 115

Brendan Gregg

JULY 4, 2021

BBR: Congestion-Based Congestion Control,” [link] 2016 - [Gregg 16] Brendan Gregg, “Unikernel Profiling: Flame Graphs from dom0,” [link] Jan 2016 - [Gregg 16b] Brendan Gregg, “Linux 4.X Titus, the Netflix container management platform, is now open source,” [link] Apr 2018 - [Cutress 19] Dr.

Performance 52

Performance Latency Hardware Storage 52

O'Reilly

APRIL 23, 2024

Google not only crawled and indexed virtually every page on the web, it looked at how sites linked to each other, tracked which of the ten top links it showed were clicked on the most, which ones led people to come back and try another and which sent them away satisfied. By the end of 2012, it was up to 82%. The market was maturing.

Social Media 108

Social Media Google Internet Internet 108

ACM Sigarch

APRIL 26, 2018

EPU: Emotion Processing Unit is designed by Emoshape , as the MCU microchip design to enable a true emotional response in AI, robots and consumer electronic devices as a result of a virtually unlimited cognitive process. FPU: Floating Processing Unit (FPU).

Hardware 99

Hardware Artificial Intelligence Architecture Google 99

Sutter's Mill

NOVEMBER 6, 2012

. … recent progress and near-future directions for C++, both at Microsoft and across the industry, and talks about some announcements related to C++11 support in VC++ 2012 and the formation of the Standard C++ Foundation. Thanks to everyone who tuned in.

C++ 40

C++ Tuning Virtualization Development 40

SQL Performance

JUNE 21, 2019

Prior to SQL Server 2012, the maximum server memory value only applied to the bufferpool, so customers needed to limit the amount of memory the bufferpool could use to save memory for the operating system and other processes. Greater than 16, keep MAXDOP at half the number of logical processors per NUMA node with a MAX value of 16.

Servers 49

Servers Azure Hardware Database 49

SQL Server According to Bob

NOVEMBER 29, 2016

What does this look like in a Virtual Machine? By their nature, virtual machines don’t have a concept of logical or physical processors. So the configuration of the virtual machine you setup and the capabilities of your hypervisor will dictate what the guest OS and SQL Server will detect. Hyper-V calls these vCPUs.

Servers 45

Servers Virtualization Scalability Hardware 45

Smashing Magazine

JUNE 30, 2021

In terms for performance, these phones have plateaued at the performance of a 2012 iPhone. 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. Nothing is slow (or fast) until you measure.

Games 145

Games Architecture Code C++ 145

Smashing Magazine

NOVEMBER 8, 2018

Note: It makes sense to use S3 as the cloud object storage solution only if we also use Amazon’s service for virtual servers on the cloud, EC2 , for running the application. Otherwise, the user may have to wait extra seconds from uploading/downloading an image to/from a distant location.

AWS 44

AWS Servers Storage Cloud 44

ScaleOut Software

MAY 15, 2014

These are areas in which we have invested heavily to take advantage of 10 Gbps (and faster) networks and to handle intermittent network delays inherent in virtual server infrastructures. Also, since data is hosted in managed code, access delays due to garbage collection (as well as the exceptions noted above) are to be expected.

Cache 40

Cache Servers Network Design 40

The Netflix TechBlog

AUGUST 29, 2023

These design principles led us to client-side load-balancing, and the 2012 Christmas Eve outage solidified this decision even further. The abstractions that Eureka provides for this are Virtual IPs (VIPs) for insecure communication, and Secure VIPs (SVIPs) for secure.

Traffic 220

Traffic Latency Cloud C++ 220

Smashing Magazine

OCTOBER 6, 2021

If you use CodePen, the default reset is Normalize.css , authored by Nicolas Gallagher and initially released in 2012. URL bar and virtual keyboard), and will not change its value even as the browser chrome changes. It can be found in many, many projects and is quite an opinionated reset.

Design 145

Design Availability Mobile Website 145

All Things Distributed

OCTOBER 2, 2017

With these goals in mind, In January, 2012 we launched Amazon DynamoDB , our cloud-based NoSQL database service designed from the ground up to support extreme scale, with the security, availability, performance and manageability needed to run mission-critical workloads. Manageable – The service would need to be easy to manage and operate.

Internet 128

Internet Internet AWS Performance 128

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. Werner Vogels weblog on building scalable and robust distributed systems. Comments ().

Scalability 167

Scalability Database Ecommerce Latency 167