Page Operations
Browse Space
Description
This page describes efforts associated with the collaboration tools during the 'Aluminum' development cycle. During this development cycle, a machine for the scheduling, tracking and provisioning of production systems will be configured. There will also be refinement and extension of backup and monitoring capabilities, establishing of domain control using the optiputer DNS, and a metered practice runs of data migration.
Task Outline
| Task | Priority | Description |
| Deploy Fabric on the controller machine | High | Python deployment tool |
| Move tracking repository to controller machine | High | Currently on amoeba; move to papa |
| Web page for the tracking repository | High | Basic webpage to display all information being tracked about each component of the collaboration tools |
| Backup scripts | High | Backup scripts for the UCSD based deployment. In the cloud, the scripts are applicable but EBS snapshots are more suitable |
| Scheduling and coordination of backup scripts in the cloud | High | Tom Control of the backup happens from our environment here. |
| Refining the probes | High | List of all probes against all applications on Confluence |
| Update the Intermapper map based on changes in the execution environment | Medium | Crash of a machine will require a change of the Intermapper map, because a new instance will have a new host name. |
| Plan the data verification | High | Plan the application data, etc. The checksums on all content files are the same. find + md5sum |
| Metered practice run of data migration | High | Time the performance of the data xfer from apps at UCSD to EC2. |
| Verify the data migration | High | Validate results |
| Purchase reserved instances | High | Determine what EC2 instance types should be purchased (small, large, xlarge), the region they will be deployed in (us-east-1a-c) and the duration of the lease. Alex: provide recommendation of what to purchase. Tom: provide recommendation for operation commitments. Goal: reasonable 75% resource purchase |
| Demo production system to Anthony | High | Prepare and document demo, practice and then do it |
| Switch collaboration tools to cloud | TBD | Collaboration Tools on EC2 production ready. What means production ready? |
| Do the production data migration | High | Migrate latest data from UCSD->EC2 |
| Switch the domain | High | Change nameserver records to point to EC2/UCSD/etc |
Deliverables
1. Central scheduling, tracking and provisioning machine was configured. This machine was configured for the following:
- provision and contextualize server instances
- track server information regarding each component of the collaboration tools
- provide an interface for which to view and make changes to the tracking repository
- schedule routine snapshots of ebs volumes and backups of application data
2. Results of the metered practice data migration: Metered Data Migration
3. Documentation on available monitoring probes: Intermapper Probes
4. Fully functioning and demonstrable collaboration tools