NSF grant awarded. Thus, this project is a "go" as of August 2013.
Notes for Hoffmann cluster researchers
Recap:
During our recent test period of Sol in November, researchers confirmed that the new system's hardware and base configuration fully enables the proper functioning of all the software they currently depend on within Helios.
However, the testing also uncovered important information that we could not have learned without this testing investment, which in turn will contribute to a more robust setup for researchers. For example, Lulu has installed the Intel compiler in the /software/ directory, with the proper, legal licensing.
Upcoming to do's for Hoffmann researchers, as of 12/9/13:
(1) As soon as possible, ensure researchers' temporary, test user data on Sol can be completely deleted.
- ChemIT will be deleting the content within everyone's test home directories on Sol. (They were originally copied there as a testing convenience during our test phase.)
- We do this deletion before the final copy to ensure a clean import of researchers' current production home directories from Helios.
-
- Lulu advised researchers of this step before the Thanksgiving break, so hopefully no surprises.
Bottom line: If there are any files on Sol which researchers cannot afford to lose (hopefully not- it was in test mode, and it is not being backed up!), researchers must move that data to Helios right away.
2) On a predetermined date, researchers will stop having access to Sol until it goes into production, replacing Helios.
- Tasks for ChemIT to do:
- Get the server's versioning hard disk recognized within the new head node.
- Ensure central software is installed properly and legal.
3) On a predetermined date, researchers must plan for 1-2 days of downtime for Helios (and before Sol is available).
Sol will still be unavailable to researchers. Thus, during this period when Helios, too, is turned off (to allow for its final set up), both Sol and Helios will be off-line to researchers.
We obviously must coordinate this cut-over date to ensure no loss of researcher's production data and to minimize the inconvenience of the downtime
- Tasks for ChemIT to do:
- Copy researchers' production home directory data from Helios to Sol.
- Add the older compute nodes from Helios, making Sol more powerful.
- Testing by ChemIT.
4) About 1-2 days after Helios's shut down (3, above), Sol will be again be available and in production capability to researchers.
After Sol is in production, researchers must confirm all is working as expected. And they should improve their software installations.
Tasks for ChemIT:
- Once Hoffmann researchers confirm Sol is "good" (that is, there is no need to "roll back" to using Helios due to some failure), Helios's head node can be added as a compute node to Sol.
Tasks for Hoffmann group:
- Hoffmann group members should install their shared group software in the /data/ Hoffmann/bin/ directory. Including:
- Materials Studio software. (Who? If Prasad is not available to do this, let's discuss your options- thank you.)
- vasp
- gulp
- etc.
Thank you! -ChemIT
Older notes:
Next steps
- Meet to review all options and confirm desired direction and expected timing.
- Review resources. Huayun Gen has cluster management experience, including set-up.
Draft idea
- Create a stand-along cluster using new hardware ($25K for minimum of 3 years operations (to confirm!). Thus, ~$8K/yr in hardware)).
- Uses new OS and related cluster management software.
- Install and configure necessary applications.
- Enable NetID-based access, if possible (limit 2-3 days for a "go/no-go" decision on this functionality)
- Confirm old nodes can successfully be added to that new cluster.
- Migrate users and data to new cluster.
- Migrate old nodes to new cluster.
Unknowns
- Time for install of all necessary applications, many of which are new to Lulu. Then configure, verify, and de-bug new-installation-related issues.
- Whether NetID-based access will succeed. But note that this is not a do-or-die step, thus we will limit the duration of our investigation, with the hope that we can make this happen.
Tasks and estimated timing
Top Level Task Description |
Effort Est. |
Assignee |
---|---|---|
Planning |
|
|
Discovery/ Overview mtg |
1.5 hrs |
|
Vet options and conduct needs analysis to match to hardware order |
1-2 weeks |
|
Specify exactly the systems to order within budget. Includes iterating with vendor experts. |
1 week |
|
Approval |
0 days |
|
Order & Installation |
|
|
Place & Process order |
1/2 week |
|
Delivery, after order is placed at Cornell |
~3 weeks |
|
Receive order and set-up hardware in 248 Baker Lab |
1 week |
|
Build New Cluster |
|
|
Get head node and 1st cluster node operational with OS and cluster management software |
3 weeks |
|
Test / Verify / Approval |
1 week |
|
Convert Old Cluster |
|
|
Move user accounts and data; test, prep, and do |
1 week |
|
Move old nodes to new cluster |
1 week |
|
- Lulu becomes available ~mid-September or early Oct, as of 8/21/13.
- See unknowns, above, which related to tasks which will obviously take additional time to accomplish.
Other provisioning models and related ideas
- We can walk through rates and scenarios, as appropriate.
- We can meet with CAC since they may be willing to do more with a commitment of $25K than is published with their $400 min. offering.
- Brainstorming idea: Would they be willing to add hardware to CAC's RedCloud to get a buyer of that hardware a better cost and/or privileged access?
Buy cycles, on demand
Good for irregular high-performance demands, especially if have high peaks of need and long-lasting jobs.
- Buy cycles from CAC (RedCloud, minimum of $400 for 8585 core*hour
- http://www.cac.cornell.edu/RedCloud/start.aspx
- 12 cores available at any one time on one system.
- Can access more than one system at a time, but systems are not linked.
- $400 (minimum) buys you 8585 core*hours
- This comes out to ~1 core for an entire year, non-stop.
- For 96 cores, that's $38.4K for 1 year, non-stop.(They have a max of 96 cores <http://www.it.cornell.edu/about/projects/virtual/test.cfm>.)
- 96 = 8 nodes, each with dual 6-core procs => 8 * 12 = 96
- Or, for $25K, that's ~536,562 cores*hours.
- $25K = $400*62.5 units. And each unit is 8585 core*hours, so 62.5 of them gets you 536,562.5 cores*hours.
- That comes to ~178,854 core*hours/ yr for 3 years. Which is 20.8 core system running non-stop each year. (Compare to one hardware node, which has 12 cores.)
- CNF, w/ Derek Stuart.
- A very reliable cluster, per Roald.
- Determine costs, processes, and trade-offs if use another cloud service, such as:
- Amazon. Amazon AC3?
- Google. Google Compute?
- Microsoft. Microsoft Azure?
Host hardware at CAC rather than with ChemIT
Hosting costs at CAC is for basic: Expert initial configuration, then keep the system current, and keep the lights running. Other service charged hourly.
Per the above rate calculator, the rate for 9 nodes (1 head node + 8 compute nodes) would be $8,291/yr. Or, $24,873 for 3 years for this service.
At current ChemIT rates, 9 nodes would be $321.84/yr. Or, $965.52 for 3 years of service.
- ChemIT rates are set by the CCB Computing Cmt and may change at any time. The rate for a group's single system (in a cluster or not) is $2.98/month, or $35.76/yr.
Table, related to our options
Option ==> |
ChemIT |
CAC: |
CAC: |
Amazon (EC3?) or |
Other ideas? |
---|---|---|---|---|---|
Hardware costs |
$25K |
- |
$25K |
- |
|
Hardware support |
Yes. |
- |
Yes. |
- |
|
OS install and configuration |
Yes. CentOS 6.4 |
|
Yes. CentOS 6.4 |
|
|
Cluster and queuing management |
Yes. Warewulf, with options |
- |
Yes. ROCKS, no options. |
- |
|
Research software install and configuration |
Yes |
No |
Yes; additional cost |
No |
|
Application debugging and optimization support |
Not usually. |
Yes; additional cost |
Yes; additional cost |
No. |
|