SML: Systems

Content

• Hardware

  • Processor, RAM, buses, GPU, disk, SSD, network, switches, racks, server centers

  • Bandwidth, latency and faults

• Basic parallelization paradigms

  • Trees, stars, rings, queues

  • Hashing (consistent, proportional)

  • Distributed hash tables and P2P

• Storage

  • RAID

  • Google File System / HadoopFS

  • Distributed (key, value) storage

• Processing

  • MapReduce

  • Dryad

  • S4 / stream processing

• Structured access beyond SQL

  • BigTable

  • Cassandra

Supplementary material

Slides in PDF and Keynote. If you want to extract the equations from the slides you can do so by using LaTeXit, simply by dragging the equation images into it. Scribe's notes will be available once they're ready.

Links

• Consistent hashing (Karger et al.) paper

• Stateless Proportional Caching (Chawla et al.) paper, slides

• Pastry P2P routing (Rowstron and Druschel) paper, site

• MapReduce (Dean and Ghemawat) site

• Google File System (Ghemawat, Gobioff, Leung) site

• Amazon Dynamo (deCandia et al.) slides, paper, Peter Vogel's DynamoDB post

• BigTable (Chang et al.) site

• CEPH filesystem (proportional hashing, file system) homepage, paper on distribution protocol

• CPUS Sandy Bridge, ARM micoarchitecture

• NVIDIA CUDA CUDA site

• ATI Stream Computing site

• Microsoft Dryad (Isard et al.) site

• Yahoo S4 (Neumayer et al.) site, slides, paper

• Memcached site

• Linked.In Voldemort (key,value) storage design description

• PNUTS distributed storage (Cooper et al.) paper

• SSDs (solid state drives) benchmarks

• All Things Distributed, Peter Vogel's blog

Videos

The quality is rather rough (this is unedited video straight from a Lumix GF2 with a 14mm lens which should explain the sound and the exaggerated hands). But it should help as a supplement with the slides.