All Things Distributed

MARCH 19, 2018

In machine learning, more is usually more. For example, training on more data means more accurate models. At AWS, we continue to strive to enable builders to build cutting-edge technologies faster in a secure, reliable, and scalable fashion. Machine learning is one such transformational technology that is top of mind not only for CIOs and CEOs, but also developers and data scientists.

Scalability 146

Scalability Hardware AWS Tuning 146

Uber Engineering

MARCH 22, 2018

Much of Uber’s business involves connecting people with people, making the reliability of our customer platform crucial to our success. The customer platform supports everything from ridesharing and Uber Eats , to Uber Freight and Uber for Business. Our … The post Introducing QALM, Uber’s QoS Load Management Framework appeared first on Uber Engineering Blog.

Engineering 81

Engineering Infrastructure Architecture 81

Tim Kadlec

MARCH 19, 2018

AMP has caused quite the stir from a philosophical perspective, but the technology hasn’t received as close of a look. A few weeks ago, Ferdy Christant wrote about the unfair advantage being given to AMP content through preloading. This got me wondering: how well does AMP really perform. I’ve seen folks, like Ferdy, analyze one or two pages, but I hadn’t seen anything looking at the broader picture…yet.

Cache 79

Cache Google Speed Metrics 79

Brendan Gregg

MARCH 22, 2018

TCP tracepoints have arrived in Linux! Linux 4.15 added five of them, and 4.16 (not quite released yet) added two more ( tcp:tcp_probe , and sock:inet_sock_set_state – a socket tracepoint that can be used for TCP analysis). We now have: # perf list 'tcp:*' 'sock:inet*'. List of pre-defined events (to be used in -e): tcp:tcp_destroy_sock [Tracepoint event] tcp:tcp_probe [Tracepoint event] tcp:tcp_receive_reset [Tracepoint event] tcp:tcp_retransmit_skb [Tracepoint event] tcp:tcp_retransmit_s

C++ 72

C++ Network Programming Open Source 72

All Things Distributed

MARCH 19, 2018

In machine learning, more is usually more. For example, training on more data means more accurate models. At AWS, we continue to strive to enable builders to build cutting-edge technologies faster in a secure, reliable, and scalable fashion. Machine learning is one such transformational technology that is top of mind not only for CIOs and CEOs, but also developers and data scientists.

Scalability 133

Scalability AWS Technology Technology 133

The Polyglot Developer

MARCH 21, 2018

About a month ago I had written about creating and importing private keys as well as generating public addresses for Bitcoin and several other cryptocurrencies using the Go programming language. This previous tutorial had more of an emphasis on creating cryptocurrency wallets with Golang than anything. The next step in making Bitcoin and other cryptocurrencies useful is to be able to transfer or send them to other people.

Programming 64

Programming Processing Development 64

Tim Kadlec

MARCH 22, 2018

2012 was a dark time for responsive images. Standards work had begun, but there was no consensus and many angry people (myself included). Some remained unconvinced that responsive images were even a problem that needed to be solved. But there was still a glimmer of hope in these dark days. In July of that year, Daan Jobsis discovered a technique that the Filament Group would later dub “compressive images”.

Network 76

Network Availability Performance Development 76

Uber Engineering

MARCH 22, 2018

Much of Uber’s business involves connecting people with people, making the reliability of our customer platform crucial to our success. The customer platform supports everything from ridesharing and Uber Eats , to Uber Freight and Uber for Business. Our … The post Introducing QALM, Uber’s QoS Load Management Framework appeared first on Uber Engineering Blog.

Engineering 40

Engineering Infrastructure Architecture 40

Brendan Gregg

MARCH 21, 2018

TCP tracepoints have arrived in Linux! Linux 4.15 added five of them, and 4.16 (not quite released yet) added two more ( tcp:tcp_probe , and sock:inet_sock_set_state – a socket tracepoint that can be used for TCP analysis). We now have: # perf list 'tcp:*' 'sock:inet*'. List of pre-defined events (to be used in -e): tcp:tcp_destroy_sock [Tracepoint event] tcp:tcp_probe [Tracepoint event] tcp:tcp_receive_reset [Tracepoint event] tcp:tcp_retransmit_skb [Tracepoint event] tcp:tcp_retransmit_s

C++ 40

C++ Programming Open Source Network 40

Rigor

MARCH 20, 2018

I recently was asked the following question by an online retailer: “Why should I invest in monitoring the user experience when I already have monitoring for our database, infrastructure, app server, and network?”. In my head I am thinking, “How have you allocated so many resources to monitoring your tech stack, but not your actual customer experience??”.

Monitoring 40

Monitoring Retail Ecommerce Infrastructure 40

SQL Server According to Bob

MARCH 18, 2018

What does improving resource usage on smaller machines really mean? [link]. SQL Server is designed to scale across the spectrum of small machines to enterprise class servers. The changes made in SQL Server 2017 CU4 improve resource usage on smaller machines, such as a VM or Container used for testing and development. SQL Server always strives to provide you with the best experience, “out of the box” and will continue improvements in future releases.

Servers 40

Servers Best Practices Software Engineering Operating System 40