Skip to content

Instantly share code, notes, and snippets.

@meteorfox

meteorfox/gist:f62d8e3ba7775722d2f5 Secret

Last active August 29, 2015 14:19
Show Gist options
  • Star 0 You must be signed in to star a gist
  • Fork 0 You must be signed in to fork a gist
  • Save meteorfox/f62d8e3ba7775722d2f5 to your computer and use it in GitHub Desktop.
Save meteorfox/f62d8e3ba7775722d2f5 to your computer and use it in GitHub Desktop.
Download ZIP
Raw
gistfile1.md

Compute (Nova)

Scenarios

  • Boot servers (always includes security groups)
  • Boot servers with base image
  • Boot servers with image snapshot
  • Boot servers with additional volumes
  • Boot servers with isolated networks
  • Boot server from volume
    • Unprepared volume
    • Prepared volume
  • Build, List, Delete servers at scale (more than 100 server instances)
  • Cluster builds and deletes
  • Live migration
    • Single
    • Cluster

Parameters

  • OS
  • Image size
  • Volume size
  • Flavor type
  • Number of servers

Compute KPIs

  • Build times (in seconds)
  • Max concurrent builds with acceptable performance and low failure rate
  • Volume preparation time (in seconds)
  • Migration time (in seconds)
  • Max number of concurrent migrations
  • Max number of ACTIVE servers in the cloud while sustaining acceptable performance and low failure rate.
  • Response times of list/delete servers at scale (more than 100 server instances) (in milliseconds)
  • Cloud Controller, Message Queue, DBs servers utilization metrics (CPU, Memory, Net, Storage IO, Storage Capacity)

Images (Glance)

Scenarios

  • Uploading Images
  • Deleting Images
  • List images as number of images grow
  • Image Snapshots
    • Different image sizes
  • Concurrent image snapshots

Parameters

  • Image size (Uncompressed and compressed)
  • Image format (QCOW2, RAW, etc)
  • Number of images
  • Backend (Swift)

Images KPIs

  • Upload time in seconds
  • Delete time in seconds
  • Response times of list/delete images at scale (more than 100 images including snapshots) in milliseconds
  • Glance server node and DB node server utilization metrics
  • Max number of images in the Glance servers while sustaining acceptable response time

Identity (Keystone)

Scenarios

  • Authentication
  • Token validation at scale (more than 50 tenants, each w/ 2 users)
  • Create and delete tenants
  • List tenants as number of tenants grow
  • Create and delete users
  • List users as number of users grow

Parameters

  • Backend choice (LDAP, PAM, RDBMS, etc)
  • Number of users per tenants
  • Number of tenants

Identity KPIs

  • Response times of list/create tenant/users at scale (more than 50 tenants with 2 users each)
  • Throughput of Identity service while keeping acceptable performance

Networking (Neutron)

  • Create and delete Networks, Subnets, Ports, Routers
  • Large number of tenant networks and ports
  • List Networks, Subnets, Ports, Routers

Block Storage (Cinder)

  • Concurrent create volumes
  • Concurrent delete volumes
  • List volumes at scale more than 100 volumes
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment