AWS, Cache and Java - Technology Performance Pulse

Seeing through hardware counters: a journey to threefold performance increase

The Netflix TechBlog

NOVEMBER 9, 2022

We decided to move one of our Java microservices?—?let’s to a larger AWS instance size, from m5.4xl (16 vCPUs) to m5.12xl (48 vCPUs). We turned to JVM-specific profiling, starting with the basic hotspot stats, and then switching to more detailed JFR (Java Flight Recorder) captures to compare the distribution of the events.

Hardware

Hardware Cache Performance Latency

Kubernetes in the wild report 2023

Dynatrace

JANUARY 16, 2023

Java, Go, and Node.js Most Kubernetes clusters in the cloud (73%) are built on top of managed distributions from the hyperscalers like AWS Elastic Kubernetes Service (EKS), Azure Kubernetes Service (AKS), or Google Kubernetes Engine (GKE). Java, Go, and Node.js Java Virtual Machine (JVM)-based languages are predominant.

Open Source

Open Source Java Operating System Programming

Migrating Netflix to GraphQL Safely

The Netflix TechBlog

JUNE 14, 2023

A single API team maintained both the Java implementation of the Falcor framework and the API Server. And we definitely couldn’t replay test non-functional requirements like caching and logging user interaction. Watch our Chaos Engineering talk from AWS Reinvent to learn more about Sticky Canaries.

Traffic

Traffic Latency Metrics Cache

Remote Workstations for the Discerning Artists

The Netflix TechBlog

MARCH 8, 2021

Below is a broad technical overview of how to go from an AWS instance to a Netflix Workstation. Instead, we created a service to take the most popular configurations and cache them. A gRPC Java Spring Boot control plane and a Golang agent manages and reports on the lifecycle. Now that you know why, here is how we did it.

Entertainment

Entertainment Storage Open Source Hardware

Elasticsearch Indexing Strategy in Asset Management Platform (AMP)

The Netflix TechBlog

MARCH 10, 2023

No matter whether you use in-house deployments or hosted solutions, you can quickly stand up an Elasticsearch cluster, and start integrating it from your application using one of the clients provided based on your programming language (Elasticsearch has a rich set of languages it supports; Java, Python, .Net, Net, Ruby, Perl etc.).

Strategy

Strategy Cache Storage Analytics

How Netflix microservices tackle dataset pub-sub

The Netflix TechBlog

OCTOBER 16, 2019

Often the data is held in memory by consumers and used as a “total cache”, where it is accessed at runtime by client code and atomically swapped out under the hood. for example Open Connect Appliance cache configuration, supported device type IDs, supported payment method metadata, and A/B test configuration.

Cache

Cache Architecture Metrics Java

Embrace event-driven computing: Amazon expands DynamoDB with streams, cross-region replication, and database triggers

All Things Distributed

JULY 14, 2015

Streams provide you with the underlying infrastructure to create new applications, such as continuously updated free-text search indexes, caches, or other creative extensions requiring up-to-date table changes. An AWS Lambda function is a simpler option that you can use, as it only requires you to code the logic, set it, and forget it.

Database

Database Lambda AWS IoT

Netflix End of Series 1

Brendan Gregg

APRIL 14, 2022

On the Netflix Java/Linux/EC2 stack there were no working mixed-mode flame graphs, no production safe dynamic tracer, and no PMCs: All tools I used extensively for advanced performance analysis. I joined Netflix in 2014, a company at the forefront of cloud computing with an attractive [work culture].

Java

Java Cloud Cache AWS

Stuff The Internet Says On Scalability For July 20th, 2018

High Scalability

JULY 20, 2018

That means multiple data indirections mean multiple cache misses. Mark LaPedus : MRAM, a next-generation memory type, is being touted as a replacement for embedded flash and cache applications. crabbone : This is the prism through which Java programmers view the world. They are very expensive. They never question this belief.

Internet

Internet Internet Scalability Automotive

Amazon DynamoDB Accelerator (DAX): Speed Up DynamoDB Response Times from Milliseconds to Microseconds without Application Rewrite.

All Things Distributed

JUNE 21, 2017

Today, I'm excited to announce the general availability of Amazon DynamoDB Accelerator (DAX) , a fully managed, highly available, in-memory cache that can speed up DynamoDB response times from milliseconds to microseconds, even at millions of requests per second. DynamoDB was the first service at AWS to use SSD storage.

Speed

Speed Cache Latency AWS

Analyzing a High Rate of Paging

Brendan Gregg

AUGUST 29, 2021

1072-aws (xxx) 12/18/2018 _x86_64_ (16 CPU) avg-cpu: %user %nice %system %iowait %steal %idle 5.03 1072-aws (xxx) 12/18/2018 _x86_64_ (16 CPU) avg-cpu: %user %nice %system %iowait %steal %idle 5.03 Reads usually have apps waiting on them; writes may not (write-back caching). Counting cache functions. Hit Ctrl-C to end. ^C

Cache

Cache C++ AWS Java

DBLog: A Generic Change-Data-Capture Framework

The Netflix TechBlog

DECEMBER 17, 2019

There are services at Netflix that use RDBMS kind of databases such as MySQL or PostgreSQL via AWS RDS. This way, log event processing can resume event-by-event afterwards, eventually discovering the watermarks, without ever needing to cache log event entries. The destination may be a datastore or an external API.

Database

Database Traffic Transportation Open Source

DBLog: A Generic Change-Data-Capture Framework

The Netflix TechBlog

DECEMBER 17, 2019

There are services at Netflix that use RDBMS kind of databases such as MySQL or PostgreSQL via AWS RDS. This way, log event processing can resume event-by-event afterwards, eventually discovering the watermarks, without ever needing to cache log event entries. The destination may be a datastore or an external API.

Database

Database Traffic Transportation Open Source

Netflix End of Series 1

Brendan Gregg

APRIL 14, 2022

On the Netflix Java/Linux/EC2 stack there were no working mixed-mode flame graphs, no production safe dynamic tracer, and no PMCs: All tools I used extensively for advanced performance analysis. Netflix has been the best job of my career so far, and I'll miss my colleagues and the culture.

Java

Java Cloud Cache AWS

Intro to Redis Cluster Sharding – Advantages, Limitations, Deploying & Client Connections

High Scalability

FEBRUARY 19, 2019

Redis Cluster is the native sharding implementation available within Redis that allows you to automatically distribute your data across multiple nodes without having to rely on external tools and utilities. At ScaleGrid, we recently added support for Redis Clusters on our platform through our fully managed Redis hosting plans.

Availability

Availability Open Source Cache Storage

An open-source benchmark suite for microservices and their hardware-software implications for cloud & edge systems

The Morning Paper

MAY 12, 2019

The suite is built using popular OSS applications and representative technologies, deliberately using a mix of languages (C/C++, Java, Javascript, node.js, Python, Ruby, Go, Scala, …) and both RESTful and RPC (Thrift, gRPC) style service interfaces. There’s a nice nod to the Weave Sockshop microservices sample application here too.

Open Source

Open Source Hardware Benchmarking Systems

Analyzing a High Rate of Paging

Brendan Gregg

AUGUST 29, 2021

1072-aws (xxx) 12/18/2018 _x86_64_ (16 CPU) avg-cpu: %user %nice %system %iowait %steal %idle 5.03 1072-aws (xxx) 12/18/2018 _x86_64_ (16 CPU) avg-cpu: %user %nice %system %iowait %steal %idle 5.03 Reads usually have apps waiting on them; writes may not (write-back caching). This is "cache busting." Hit Ctrl-C to end. ^C

Cache

Cache C++ AWS Systems

Smashing Podcast Episode 44 With Chris Ferdinandi: Is The Web Dead?

Smashing Magazine

NOVEMBER 30, 2021

Your template goes here, your Java script goes here, your CSS goes here. That was awful to develop for. So I can have a multipage app, cache my API calls for a short period of time without having to cache them in memory. And that’s not a knock on the people behind the standards processes. Drew: It was, yeah.

Technology

Technology Technology Servers Code

Egnyte Architecture: Lessons learned in building and scaling a multi petabyte content platform

High Scalability

NOVEMBER 25, 2019

Egnyte is a secure Content Collaboration and Data Governance platform, founded in 2007 when Google drive wasn't born and AWS S3 was cost-prohibitive. Edge caching. In general, Egnyte connect architecture shards and caches data at different levels based on: Amount of data. Languages: Java. Nginx for disk based caching.

Architecture

Architecture Cache Azure Storage

Technology Performance Pulse

Seeing through hardware counters: a journey to threefold performance increase

Kubernetes in the wild report 2023

Trending Sources

Migrating Netflix to GraphQL Safely

Remote Workstations for the Discerning Artists

Elasticsearch Indexing Strategy in Asset Management Platform (AMP)

How Netflix microservices tackle dataset pub-sub

Embrace event-driven computing: Amazon expands DynamoDB with streams, cross-region replication, and database triggers

Netflix End of Series 1

Stuff The Internet Says On Scalability For July 20th, 2018

Amazon DynamoDB Accelerator (DAX): Speed Up DynamoDB Response Times from Milliseconds to Microseconds without Application Rewrite.

Analyzing a High Rate of Paging

DBLog: A Generic Change-Data-Capture Framework

DBLog: A Generic Change-Data-Capture Framework

Netflix End of Series 1

Intro to Redis Cluster Sharding – Advantages, Limitations, Deploying & Client Connections

An open-source benchmark suite for microservices and their hardware-software implications for cloud & edge systems

Analyzing a High Rate of Paging

Smashing Podcast Episode 44 With Chris Ferdinandi: Is The Web Dead?

Egnyte Architecture: Lessons learned in building and scaling a multi petabyte content platform

Stay Connected