Resumes that made it into FAANG

FAANG style resume tips

https://code.likeagirl.io/resumes-that-made-it-into-faang-f7de9b0c4396

#resume #cv #tips #faang #google #facebook #flipkart #bloomberg #microsoft #servicenow #cisco #jobs

💻 The byte is not enough

What is Hindley-Milner? (and why is it cool?)

Quick insight on one of the most popular type systems.

https://web.archive.org/web/20181118000004/http://www.codecommit.com/blog/scala/what-is-hindley-milner-and-why-is-it-cool

#hindleymilner #typesystem #scala #haskell #fp #algorithms #typetheory

💻 The byte is not enough

The Netflix Simian Army

Chaos Monkey, a tool that randomly disables production instances to make sure [we] can survive this common type of failure without any customer impact.

https://netflixtechblog.com/the-netflix-simian-army-16e57fbab116

#netflix #chaosmonkey #systemdesign #faulttolerance #cloud #computation #security

💻 The byte is not enough

How many nodes are talked to with Quorum in Cassandra? Also should I use it?

A brief note on Cassandra's consistency levels.

https://foundev.medium.com/cassandra-how-many-nodes-are-talked-to-with-quorum-also-should-i-use-it-98074e75d7d5

#facebook #cassandra #database #systemdesign #consistency #quorum #architecture #design #distributed #keyvalue #dht

💻 The byte is not enough

Consistent Hashing

High-level overview of consistent hashing algorithm first published by David Karger et al. back in 1997.

https://tom-e-white.com/2007/11/consistent-hashing.html

#systemdesign #architecture #distributed #system #hashing #consistenthashing #algorithms

💻 The byte is not enough

🧑‍🔬 Google File System (GFS)

A classical paper on Google File System dating back to 2003 devoted to the distributed file system developed by Google with the main aim to store and process enormous amounts of data at scale effectively.

The work influenced the creation of two other well-known technologies like HDFS and Google's BigTable.

https://www.youtube.com/watch?v=eRgFNW4QFDc

#systemdesign #google #distributed #filesystems #design #reliability #performance #scalability #faulttolerance #research #paper

💻 The byte is not enough

🧑‍🔬 Large-scale cluster management at Google with Borg

An incredible paper on Google's Borg, a Kubernetes predecessor, and how Google successfully managed tens of thousands of machine clusters for over a decade.

Google's Borg system is a cluster manager that runs hundreds of thousands of jobs, from many thousands of different applications, across a number of clusters each with up to tens of thousands of machines.

Watch the Borg presentation at EuroSys 2015:
https://www.youtube.com/watch?v=7MwxA4Fj2l4

#systemdesign #google #borg #distributed #orchestrator #design #reliability #performance #scalability #faulttolerance #research #paper

💻 The byte is not enough

🧑‍🔬 Consistent Hashing and Random Trees: Distributed Caching Protocols for Relieving Hot Spots on the World Wide Web

Another fundamental work on distributed hashing protocols that influenced distributed systems' evolution. Nowadays, consistent hashing is being used in many software distributions like Amazon Dynamo, Apache Cassandra, Riak, Voldemort to name a few. Few big online platforms are known to implement consistent hashing algorithms to scale for performance, availability, and reliability.

More concise take on the matter:

https://www.toptal.com/big-data/consistent-hashing

#systemsdesign #distributed #design #reliability #performance #availability #scalability #research #paper #consistent #hashing

💻 The byte is not enough

🧑‍🔬 Twine: A Unified Cluster Management System for Shared Infrastructure

If you as me were impressed by the impressive work Google done in their Borg system, and it's successor Kubernetes, then you will definitely enjoy learning how Facebook makes use of their infrastructure in an astonishing paper on Facebook Twine. This tremendous work benefits from experience gained through developing and managing other popular systems like Borg, Kubernetes or Mesos, and aims to scale to more than a million of machines.

https://engineering.fb.com/2019/06/06/data-center-engineering/twine/

#systemdesign #facebook #twine #distributed #orchestrator #design #reliability #performance #scalability #faulttolerance #research #paper #borg #kubernetes #mesos

💻 The byte is not enough

🧑‍🔬 Scaling Memcache at Facebook

While reading Facebook's Twine cluster management system white paper, I noticed an interesting thing in section 5 where paper authors claim to have a highly optimized memcached deployment that can handle "930k lookups per second on an 18-core/36-hyperthread machine". Just think for a second, 930 000 lookup requests per second on a single machine. How the heck they could achieve this kind of performance. To answer this and many other questions, I went on reading another Facebook white paper on optimizing Memcached for meeting the world's largest social network needs.

A short version listing all the big things Facebook incorporated into their version of Memcached can be found here:

https://medium.com/@shagun/scaling-memcache-at-facebook-1ba77d71c082

#systemdesign #facebook #memcached #distributed #caching #design #reliability #performance #scalability #faulttolerance #research #paper #redis #dht

💻 The byte is not enough

🧑‍🔬 TAO: Facebook’s Distributed Data Store for the Social Graph

Have you ever wondered how Facebook manages its social graph, containing petabytes of user data? What techniques do they apply to serve billions of reads and millions of writes each second? All this and much more in another great white paper on Facebook's TAO - a geographically distributed data store that provides efficient and timely access to the social graph for Facebook’s demanding workload using a fixed set of queries.

https://engineering.fb.com/2013/06/25/core-data/tao-the-power-of-the-graph/

USENIX ATC '13 - TAO: Facebook’s Distributed Data Store for the Social Graph:
https://www.youtube.com/watch?v=sNIvHttFjdI

#systemdesign #facebook #memcached #tao #socialgraph #api #caching #design #reliability #performance #scalability #faulttolerance #consistency #research #paper #mysql #infrastructure

💻 The byte is not enough

🧑‍🔬 Paxos Made Simple

In the year 1989 Leslie Lamport, a known computer scientist in the field of distributed systems, published his tremendous work on Paxos — a family of protocols for solving consensus in a network of unreliable or fallible processors. Though from the very beginning, the proposed algorithm was diminished by computer science society due to its complexity, it started gaining significant recognition after almost 10 years since first published having a second coming in 1998. In late 2001, Lamport published a simplified version of the original paper, discarding unnecessary information and providing the description on the backbone of Paxos protocol.

Paxos Simplified by Chris Colohan:
https://www.youtube.com/watch?v=SRsK-ZXTeZ0

#systemdesign #paxos #consensus #lamport #design #reliability #performance #faulttolerance #scalability #consistency #quorum #research #paper #infrastructure #distributed

💻 The byte is not enough

🧑‍🔬 MapReduce: Simplified Data Processing on Large Clusters

Another seminal work from the past that established distributed systems' evolution for decades ahead. The MapReduce model is probably the most well know programming model designed for processing and generating big data sets with a parallel, distributed algorithm on a cluster.

#systemdesign #mapreduce #design #performance #faulttolerance #scalability #research #paper #infrastructure #distributed #computation #cluster #gfs

💻 The byte is not enough

🧑‍🔬 Turbine: Facebook’s Service Management Platform
for Stream Processing

A scalable service management platform for Facebook’s stream processing service. Turbine is designed to bridge the gap between the capabilities of existing general-purpose cluster management frameworks like Tupperware and Facebook’s stream processing requirements. In production for several years now, Turbine has enabled a boom in stream processing at Facebook.

https://engineering.fb.com/2020/04/21/data-infrastructure/turbine/

#systemdesign #turbine #facebook #streaming #processing #design #performance #acid #reliability #faulttolerance #scalability #research #paper #infrastructure #distributed #cluster #management

💻 The byte is not enough

LogDevice: a distributed data store for logs

A log is the simplest way to record an ordered sequence of immutable records and store them reliably. Build a data intensive distributed service and chances are you will need a log or two somewhere. At Facebook, we build a lot of big distributed services that store and process data. Want to connect two stages of a data processing pipeline without having to worry about flow control or data loss? Have one stage write into a log and the other read from it. Maintaining an index on a large distributed database? Have the indexing service read the update log to apply all the changes in the right order. Got a sequence of work items to be executed in a specific order a week later? Write them into a log, have the consumer lag a week. Dream of distributed transactions? A log with enough capacity to order all your writes makes them possible. Durability concerns? Use a write-ahead log.

https://engineering.fb.com/2017/08/31/core-data/logdevice-a-distributed-data-store-for-logs/

#systemdesign #logdevice #log #wal #logsdb #rocksdb #consensus #paxos #quorum #lsmtree #facebook #storage #design #performance #scalability #research #faulttolerance #infrastructure

💻 The byte is not enough

Database Storage Engines: B-Tree vs LSM-Tree

Have you ever concern yourself with the question of how modern database systems' internal storage works? Well, there are two popular ways to handle data storage — a B-Tree (a generalization of Binary Search Tree) and a Log-Structured Merge Tree, both with having pros and cons.

Here are a few short articles to get a grasp of the trade-offs between the two:

1. https://blog.yugabyte.com/a-busy-developers-guide-to-database-storage-engines-the-basics/
2. https://blog.yugabyte.com/a-busy-developers-guide-to-database-storage-engines-advanced-topics

3. https://rkenmi.com/posts/b-trees-vs-lsm-trees

4. https://tikv.org/deep-dive/key-value-engine/b-tree-vs-lsm/

#systemdesign #dbs #databases #design #btree #lsmtree #storage #performance #sql #nosql #engine

💻 The byte is not enough

🧑‍🔬 Architecture of ZAB – ZooKeeper Atomic Broadcast protocol

The ZAB protocol ensures that the Zookeeper replication is done in order and is also responsible for the election of leader nodes and the restoration of any failed nodes. In a Zookeeper ecosystem, the leader node is the heart of everything; every cluster has one leader node and the rest of the nodes are followers. All incoming client requests and state changes are received at first by the leader with responsibility to replicate it across all its followers (and itself). All incoming read requests are also load balanced by the leader within itself and its followers.

Original ZAB paper:
https://marcoserafini.github.io/papers/zab.pdf

Implementation details of ZAB:
http://www.tcs.hut.fi/Studies/T-79.5001/reports/2012-deSouzaMedeiros.pdf

#systemdesign #zab #consensus #total #order #broadcast #design #quorum #performance #faulttolerance #yahoo #paper #research #scalability

💻 The byte is not enough

Building Facebook’s service encryption infrastructure

What does it take to incorporate security protocols into a system of millions of services running on thousands of machines across the world? You could use some well-known technologies as Kerberos, but will soon realize it does not work on a big scale. You would then probably stick to the idea of building a custom in-house solution, which is exactly what Facebook did to suit their needs in security and operability without compromising performance.

https://engineering.fb.com/2019/05/29/security/service-encryption/

#systemdesign #security #facebook #kerberos #tls #encryption #traffic #networking #dns #performance #faulttolerance #scalability #infrastructure

💻 The byte is not enough

🧑‍🔬 CORFU: A Distributed Shared Log

Despite almost forty years of research into replicated storage schemes, the only approach so far to scale up capacity and throughput has been to shard data and trade consistency for performance. The CORFU system breaks this seeming tradeoff by organizing a cluster of drives as a single, shared log. CORFU offers a single-copy semantics at cluster-scale speeds, providing a scalable source of atomicity and durability for distributed systems.

https://www.youtube.com/watch?v=GmVQVT9aZfU

Original paper:
https://www.cs.utexas.edu/~lorenzo/corsi/cs380d/papers/a10-balakrishnan.pdf

#systemdesign #corfu #microsoft #design #reliability #algorithms #distributed #log #replication #consensus #performance #scalability #faulttolerance #paper #research #infrastructure

💻 The byte is not enough

🧑‍🔬 Delos: Simple, flexible storage for the Facebook control plane

Facebook Delos — is a fundamentally new architecture for building replicated storage systems. Its modular, layered design provides flexibility and simplicity without sacrificing performance or reliability. Delos enables fast time-to-deployment for new storage systems — Facebook deployed an initial version in production within eight months — as well as safe, rapid evolution. Facebook swapped in a new ordering mechanism to obtain 10x lower latency without any service downtime.

https://www.youtube.com/watch?v=wd-GC_XhA2g&t=314s

Blog post presentation:
https://engineering.fb.com/2019/06/06/data-center-engineering/delos/

Original paper:
https://www.usenix.org/system/files/osdi20-balakrishnan.pdf

#systemdesign #delos #facebook #distributed #shared #log #algorithms #design #consensus #performance #scalability #faulttolerance #paper #research #infrastructure #storage #controlplane #corfu #paxos #zookeeper #zab

💻 The byte is not enough

Building a real-time user action counting system for ads

A quick read on how Pinterest built an ad tracking platform called Aperture to count the number of ad interactions in the past one, three, or 30 days.

https://medium.com/pinterest-engineering/building-a-real-time-user-action-counting-system-for-ads-88a60d9c9a

#systemdesign #pinterest #aperture #kafka #distributed #counter #infrastructure #tech #blog #adtech #ads #rocksdb

💻 The byte is not enough