Skip to end of metadata
Go to start of metadata

Jetstream can be used in several different virtual machine (VM) sizes which are charged in service units (SUs) based on how much of the total system resource is used. The basic unit of VM allocation for Jetstream is based on a virtual CPU (vCPU) hour: 1 service unit (SU) is equivalent to 1 vCPU for 1 hour of wall clock time. A standard ‘Tiny’ VM instance consists of 1 vCPU, 2 GB of RAM, and 8 GB of storage. This corresponds closely to a ‘t2.small’ instance in Amazon Web Services. The majority of storage within an instance is available for user data but will vary based on the VM image selected.  The table below outlines the VM sizes created for Jetstream.


VM Size

vCPUs

RAM (GB)

Local Storage (GB)

SU cost per hour*

Can be saved as
an image

m1.tiny

1

2

8

1

✔︎ yes

m1.small

2

4

20

2

✔︎ yes

m1.medium

6

16

60

6

✔︎ yes

m1.large

10

30

60

10

✔︎ yes

m1.xlarge

24

60

60

24

✔︎ yes

m1.xxlarge

44

120

60

44

✔︎ yes
s1.large

10

30

120

10

✘ No
s1.xlarge

24

60

240

24

✘ No
s1.xxlarge

44

120

480

44

✘ No

This allocation information may be subject to changes in the future.

Please note that s1.* based customized instances will NOT be able to be used to create images in Atmosphere.

If your work requires 24 GB of RAM and 60 GB of local storage, then you would request 10 SUs per hour to cover a single Large VM instance.

If your work requires 10 GB of local storage in 1 thread using 3 GB of RAM, then you would request 2 SUs per hour for a Small VM instance.

You would then multiply by the number of hours you will use that size VM in the next year and multiply by the number of VMs you will need.

To calculate the number of SUs you will need in the next year, first estimate the number of hours you expect to work on a particular project.  For example, if you typically work 40 hours per week and expect to spend 25% of your time on this project that would be 10 hours per week.  Next, calculate the total number of hours per year for this project:

     Total hours = 10 hours per week * 52 weeks per year

     Total hours = 520

Finally, calculate the total SUs for the year for a single instance medium VM:

     Total SUs = 520 hours per year * vCPUs

     Total SUs = 520 hours per year * 6vCPUs 

     Total SUs = 3120

If project requires more than 1 medium size VM multiply total SUs by the number of VMs that you will need:

     Total SUs needed for 3 medium size VMs = 3 * 3120

     Total SUs = 9360


* This is described in the XSEDE KB document On XSEDE, how are compute jobs charged?


Shutdown your VM properly

The calculations above assume that your VM is shutdown properly.  For instructions see Instance management actions.


For information on submitting a Research Allocation Request, please see https://portal.xsede.org/successful-requests.  Note that all allocations above the startup level require a strong justification for the time being requested.









  • No labels