This post is getting gunicorn,nginx, Django, postgresql up and running in EC2.
Obviously inspired by the Instagram stack and their scalable architecture, I decided to give the stack a try…
Fab files, Git usage, I/O optimisation for EBS, auto-scaling, memcache, redis usage and clustering are for laters but for now the aim was to get the base stack in place. Made use of various references on the web as well as from github and they are mentioned at the end of this post.
P.S: Assumpton is one knows- AWS or at least the …

Amazon’s Elastic Load Balancer
Amazon’s Route53
nginx
HAProxy use extensively
Gunicorn + Fabric for deploment
Django
vmtouch- Smartly use vmtouch for cache diagnostics and have scripts in python to dynamically generate vmtouch commands for different servers by parsing vmtouch output in one server.
PostgreSQL – shard the data and in clusters.
PostgreSQL instances run in a master-replica setup using Streaming Replication
Pgbouncer connection pooling for PostgreSQL
EBS drives in a software RAID using mdadm
XFS as the file system
bonnie++ for benchmarking EBS I/O and tuning
(I/O performance for EBS Matters – Do not replicate solutions as-is)
ec2-consistent-snapshot
repmgr from 2ndQuadrant
Amazon S3
Amazon CloudFront as …

EUCALYPTUS – Elastic Utility Computing Architecture for Linking Your Programs To Useful Systems – is an open-source software infrastructure for implementing “cloud computing” on clusters.

demaNtigrate uses Flex and it’s component library for the RIA as a browser based integration designer, with ServiceMix distribution components as it’s runtime engine.

Various SaaS and SIP providers, such as Salesforce.com, eloqua, Success Factors, Informatica, Open Air, nSite, Jamcracker are offering integration APIs, connectors