Tag: 5

NEXTPLATFORM AD

The Perfect AI Storage: Trino From Facebook And Iceberg From Netflix?

With MTIA v2 Chip, Meta Can Do AI Inference, But Not Training

NEXTPLATFORM AD

Meta Platforms Crafts Homegrown AI Inference Chip, AI Training Next

The Iron That Will Drive AI At Meta Platforms

Meta’s Velox Means Database Performance Is Not Subject To Interpretation

Meta Platforms Hacks CXL Memory Tier Into Linux

NEXTPLATFORM AD

Once Again, Meta Buys Rather Than Builds A Supercomputer

The Most Obvious Hyperscaler To Do Custom Chips Was Always Facebook

Xeon D Refresh: The Little Hyperscale Engine That Could

Surfing On The Ethernet Bandwidth Waves, Avoiding The Rocks

The Opposite Of Snowflake: Analytics Without The Data Warehouse

AMD Gets Inside Facebook’s Latest – And Most Powerful – Microserver

NEXTPLATFORM AD

Getting Meta: Abstracting And Multisourcing The Network Like An FBOSS

ML Training Growing Pains at Facebook

Vertically Unchallenged

How Facebook Architects Around Silent Data Corruption

The Once And Future Federated Database

Presto Is The Third Time Charm For Federated Databases

Open Compute Really Is Open For Business

Inside Facebook’s Future Rack And Microserver Iron

A Hackathon To Battle The Coronavirus Pandemic

Future Kubernetes Will Mimic What Facebook Already Does

How Facebook Might Find Nervana For Machine Learning Training

How To Benefit From Facebook’s New Network Fabric

NEXTPLATFORM AD

Inside Facebook’s Future Machine Learning Platforms

The Microserver, Redux And Reconsidered

Facebook Sounds Opening Bell for AI Inference Hardware Makers

Machine Learning Drives Changing Disaster Recovery At Facebook

Facebook’s Expanding Machine Learning Infrastructure

Lessons Learned From Facebook’s Split Network Backbone

NEXTPLATFORM AD

Machine Learning Gets An InfiniBand Boost With Caffe2

A Look at Facebook’s Interactive Neural Network Visualization System

Facebook Pushes The Search Envelope With GPUs

Open Hardware Pushes GPU Computing Envelope

A Peek Inside Facebook’s Server Fleet Upgrade

Microsoft Azure Goes Back To Rack Servers With Project Olympus

One Rack To Stack Them All

Facebook Flow Is An AI Factory Of The Future

Datacenters To Get A High Fiber Bandwidth Diet

Xeon D Shows How ARM Can Beat Intel

Shiny New Open Compute Gear, And The Roadmap Ahead

Inside The Systems That Drive Facebook

Microsoft’s Open Compute Gambit Pays Off

Open Hardware Means More Than Saving Money For Rackspace

The Myth Of In-Memory Computing

Facebook To Open Up Custom Machine Learning Iron

Facebook Busts Out Of Homegrown Switches

Three Is The Magic Number For Hardware Cycles

Three Secrets Of How Facebook Copes With Scale

NoSQL Pioneer Readies Disruptive Scale-Out Storage

How Facebook “Hack” is Upending Programming Efficiency

HP Follows Hyperscale Lead With Composable Infrastructure

Deep Learning Pioneer Pushing GPU Neural Network Limits

Cold Storage Heats Up At Facebook

Cassandra Carves Storage Niche, Aims At Microservers

Etsy Shows How To Be Just Crafty Enough With Systems

Facebook Ops Director On Breaking Open The Switch

Lessons And Time Drive Better Microserver Designs

How Machine Learning Will Improve Servers