Go memory ballast: How I learned to stop worrying and love the heap
I’m a big fan of small code changes that can have large impact. This may seem like an obvious thing to state, but let me explain:
These type of changes often involve diving into and understanding things one is not familiar with.
Even with the most well factored code, there is a maintenance cost to each optimization you add, and it’s usually (although not always) pretty linear with the amount of lines of code you end up adding/changing.
We recently rolled out a small change that reduced the CPU utilization of our API frontend servers at Twitch by ~30% and reduced overall 99th percentile API latency during peak load by ~45%.
This blog post is about the change, the process of finding it and explaining how it works.
Using Z-order indexing, you can efficiently run range queries on any combination of fields in your schema. Although Amazon DynamoDB doesn’t natively support Z-order indexing, you can implement the functionality entirely from the client side. A single Z-order index can outperform and even replace entire collections of secondary indexes, saving you money and improving your throughput.
In a previous AWS Database Blog post, I introduced Z-order indexing, a way in which you can sort your data to efficiently query an Amazon DynamoDB table by using range bounds on multiple attributes. In this post, we explore the process of creating a schema for your index. We look at how to decide which attributes to include in your schema, how your index’s schema impacts query efficiency, and how to work with a variety of data types.
This post builds on concepts that are described in Part 1, so I recommend taking some time to review it before diving in.
As custodians and builders of the streaming platform at Grab operating at massive scale (think terabytes of data ingress each hour), the Coban team’s mission is to provide a NoOps, managed platform for seamless, secure access to event streams in real-time, for every team at Grab.
Streaming systems are often at the heart of event-driven architectures, and what starts as a need for a simple message bus for asynchronous processing of events quickly evolves into one that requires a more sophisticated stream processing paradigms. Earlier this year, we saw common patterns of event processing emerge across our Go backend ecosystem, including:
Filtering and mapping stream events of one type to another
Aggregating events into time windows and materializing them back to the event log or to various types of transactional and analytics databases
Generally, a class of problems surfaced which could be elegantly solved through an event sourcing1 platform with a stream processing framework built over it, similar to the Keystone platform at Netflix2.
This article details our journey building and deploying an event sourcing platform in Go, building a stream processing framework over it, and then scaling it (reliably and efficiently) to service over 300 billion events a week.
To simulate a production scenario, the model is trained using an example dataset containing images of an open-source printed circuit board, with defects and without. An accompanying AWS Serverless Application Repository application deploys the Step Functions workflow for handling image classification and notifications.
Uber is developing a payment platform for India that enables operations teams to more seamlessly collect and distribute cash and digital wallet payments to drivers. In this article, San Francisco-based software engineer Yijun Liu reflects on his experiences working with the Uber India Engineering team in Bangalore to architect this revamped payment system.
When it comes to timeouts, there are two types of people: those who know how tricky they can be, and those who are yet to find out.
As tricky as they are, timeouts are a reality in the connected world we live in. As I am writing this, on the other side of the table, two persons are typing on their smartphones, probably chatting to people very far from them. All made possible because of networks.
Networks and all their intricacies are here to stay, and we, who write servers for the web, have to know how to use them efficiently and guard against their deficiencies.
Without further ado, let’s look at timeouts and how they affect our net/http servers.