WIP: Roihu accounts

docs/accounts/billing.md

@@ -11,7 +11,7 @@ See the [Billing Unit and price calculator](https://research.csc.fi/billing-unit
 
 **The billing rates for the services are as follows:**
 
-* [Puhti and Mahti billing](../computing/hpc-billing.md)
+* [Roihu, Puhti and Mahti billing](../computing/hpc-billing.md)
 * Allas object storage billing:  1 TiB consumes **1.05** Storage BU per hour. Only actual
    data stored in Allas is billed.
 * [Pouta billing](../cloud/pouta/vm-flavors-and-billing.md)
@@ -31,7 +31,7 @@ See the [Billing Unit and price calculator](https://research.csc.fi/billing-unit
 In the _My Projects_ page in [MyCSC](https://my.csc.fi) you can study the
 Billing Unit consumption and apply for more Billing Units. There you can easily
 check who consumed the Billing Units, when they were consumed and in which
-service. Note that storage-related Billing Unit consumption (in Puhti/Mahti Scratch folders and in
+service. Note that storage-related Billing Unit consumption (in Roihu/Puhti/Mahti Scratch folders and in
 Allas) is not linked to a specific user account and is reported as "other" or "system".
 
 
@@ -44,7 +44,7 @@ You will be notified before your CSC Project's Billing Units run out. When an ac
 
 Detailed information on how services are currently limiting the use when Billing Units run out:
 
-* [Puhti and Mahti](../computing/usage-policy.md#running-out-of-billing-units)
+* [Roihu, Puhti and Mahti](../computing/usage-policy.md#running-out-of-billing-units)
 * [Sensitive Data Desktop](../data/sensitive-data/sd-billing-units.md#what-happens-if-your-project-runs-out-of-billing-units)
 
 

docs/accounts/how-to-access-roihu-large-partition.md

@@ -0,0 +1,86 @@
+# How to get access to Roihu large partition
+
+Projects running well-scaling codes can get access to the large partition
+(20-200 nodes) on Rahti in three steps. First, a 30-day test period for the
+large partition is requested. Second, during the test period, the scalability
+and parallel performance of the code is demonstrated with appropriate test
+runs. Finally, the results are submitted for evaluation by the project manager.
+
+The process is described in detail below.
+
+## Test access to the large partition on Roihu
+
+To request the 30-day test period, proceed as follows:
+
+1. Login to [MyCSC](https://my.csc.fi) and in the _Projects_ menu select the
+   project you want to modify.
+2. In the _Services_ list, click open the settings for **Roihu** service
+   (_Configure_). This opens a page where the project manager can modify the
+   settings for disk quotas (_Quota settings_) and request access to the large
+   partition (_Large partition settings_). Click open
+   _Large partition settings_.
+3. Click the _Apply for trial access_ button. After the access has been
+   granted, you will be able to submit jobs to the large partition.
+
+## Scalability testing
+
+In the second phase, test runs demonstrating the scalability are to be
+performed. Here are some general guidelines for scalability testing.
+
+* Testing should be done for at least three different node counts up to the
+  target in production (for example with 20, 40, 60, and 80 nodes).
+* Tests are run through the batch job system.
+* The test runs should reflect real production runs, i.e. the number of atoms,
+  number of grid points, disk I/O load etc. should be similar.
+* The input data set must be the same for each run.
+* The running time in tests should be reduced as much possible, for example, by
+  running only few time steps, iterations etc.
+* The running time should still be long enough that initialization does not
+  affect results. Typically, a few minutes for the shortest run time (largest
+  node count) is fine.
+* Parameters affecting the scalability can, and are encouraged to be, changed.
+  Note also the
+  [performance checklist](../computing/running/performance-checklist.md).
+* Minimum requirement is 75 % parallel efficiency (i.e. speedup of 1.5 when
+  doubling the number of nodes).
+
+## Reporting
+
+The scalability report should contain a short description of the software and
+the test case, as well as wall-times for each node count. If the software is
+not pre-installed by CSC, describe briefly also the parallelization strategy
+used in the software and include details about the I/O implementation and load.
+
+Attach to the report a representative batch job script, and if the application
+was run with hybrid MPI/OpenMP parallelization, attach also the `stderr` of a
+single run where the following settings are applied:
+
+```bash
+export OMP_AFFINITY_FORMAT="Process %P level %L thread %0.3n affinity %A"
+export OMP_DISPLAY_AFFINITY=true
+```
+
+Reporting the results of the test runs or applicable previous scalability data
+is done through [MyCSC portal](https://my.csc.fi) as follows:
+
+1. Login to [MyCSC](https://my.csc.fi) and in the _Projects_ menu select the
+   project you want to modify.
+2. In the _Services_ list, click open the settings for **Roihu** service
+   (_Configure_). This opens a page where the project manager can modify the
+   settings for disk quotas (_Quota settings_) and request access to the large
+   partition (_Large partition settings_). Click open
+   _Large partition settings_.
+3. For the results and justification, there is a text box and possibility to
+   attach documents (please remember to upload the documents after you have
+   selected them). Multiple documents can be attached. Finally, submit the
+   justification.
+4. CSC experts will evaluate the results and grant production access to the
+   large partition. If there is a problem with the code performance, the
+   project manager will be contacted.
+
+## Assistance
+
+CSC's experts can help users in performing scalability tests if needed and also
+provide hints for improving the performance of their software.
+[Contact CSC Service Desk](../support/contact.md) if you need assistance with
+your software.

docs/accounts/how-to-add-service-access-for-project.md

@@ -2,7 +2,7 @@
 
 Access can be granted to the following services:
 
-* Supercomputers **Puhti** and **Mahti**
+* Supercomputers **Roihu**, **Puhti** and **Mahti**
 * Cloud services **cPouta**, **Rahti** and **Pukki DBaaS**
 * Storage services **Allas** and **IDA**
 * Sensitive data services **SD Services**

docs/accounts/how-to-increase-disk-quotas.md

@@ -1,22 +1,22 @@
 # How to increase disk quotas
 
-## Increasing the storage capacity in Puhti and Mahti
+## Increasing the storage capacity in Roihu and Mahti
 
-The quotas of the [scratch](../computing/disk.md#scratch-directory) and [projappl](../computing/disk.md#projappl-directory) directories of Puhti and Mahti servers can be
+The quotas of the [scratch](../computing/disk.md#scratch-directory) and [projappl](../computing/disk.md#projappl-directory) directories of Roihu and Mahti servers can be
 increased if needed.
 
 You can use MyCSC to manage these quotas.
 
 1. Login to [MyCSC](https://my.csc.fi) and **select the project** you want to
    modify.
-1. In the Services list, click Configuration for **Puhti** or **Mahti** service as needed. This opens a page where the **project manager** can modify the quotas.
+1. In the Services list, click Configuration for **Roihu** or **Mahti** service as needed. This opens a page where the **project manager** can modify the quotas.
 
 You can check the default limits of scratch and projappl areas from the table below. Values in parenthesis indicate automatically approved limits. Applications with higher values will be determined in the resource allocation meetings every three weeks.
 
 | Directory | Default size | Maximum size | Default file number limit | Maximum file number limit |
 |-----------|--------------|--------------|---------------------------|---------------------------|
-| projappl&nbsp;(Puhti)  |   50 GiB     |  200&nbsp;GiB&nbsp;(<&nbsp;100&nbsp;GiB)    | 100 000                   | 2 000 000&nbsp;(<&nbsp;500&nbsp;000)                 |
-| scratch&nbsp;(Puhti)   |   1 TiB      |  100&nbsp;TiB&nbsp;(<&nbsp;20&nbsp;TiB)    | 1 000 000                 | 10&nbsp;000&nbsp;000 (<&nbsp;5&nbsp;000&nbsp;000)               |
+| projappl&nbsp;(Roihu)  |   15 GiB     |  250&nbsp;GiB&nbsp;(<&nbsp;100&nbsp;GiB)    | 150 000                   | 2 500 000&nbsp;(<&nbsp;1&nbsp;000&nbsp;000)                 |
+| scratch&nbsp;(Roihu)   |   250 GiB      |  100&nbsp;TiB&nbsp;(<&nbsp;10&nbsp;TiB)    | 500 000                 | 10&nbsp;000&nbsp;000 (<&nbsp;2&nbsp;500&nbsp;000)               |
 | projappl&nbsp;(Mahti)  |   50 GiB     |  200&nbsp;GiB&nbsp;(<&nbsp;100&nbsp;GiB)    | 100 000                   | 2 000 000 (<&nbsp;500&nbsp;000)                 |
 | scratch&nbsp;(Mahti)   |   1 TiB      |  200&nbsp;TiB&nbsp;(<&nbsp;40&nbsp;TiB)    | 1 000 000                 | 20&nbsp;000&nbsp;000&nbsp;(<&nbsp;10&nbsp;000&nbsp;000)               |
 

docs/accounts/how-to-manage-your-project.md

@@ -48,7 +48,7 @@ Users are not able to use the services which have been applied for the Project a
 | Service | What happens right after Project is closed |
 | -- | -- |
 | Allas, Fairdata IDA, SD Connect | Project is disabled, and users will not be able to use the service in the context of the Project. |
-| Mahti and Puhti | No new Slurm jobs can be submitted. Users cannot access the Project data anymore. |
+| Mahti and Roihu | No new Slurm jobs can be submitted. Users cannot access the Project data anymore. |
 | cPouta and ePouta | Project is disabled, and users will not be able to use the service in the context of the Project. All virtual machines of the Project will be powered off. |
 | Rahti | Project pod quota is set to 0 and all workloads are stopped. |
 | SD Desktop | Desktops are paused, Project can't launch new desktops. |

docs/accounts/strong-identification.md

@@ -3,10 +3,9 @@
 Using strong identification means that you prove your identity. After you have been authenticated using a strong authentication method, 
 your assurance level is raised from low to medium. Some of CSC's services require a higher level of assurance. 
 
-Sensitive Data service is one of them:
-[Sensitive Data](../data/sensitive-data/findata-permit.md)
+Accessing Roihu requires medium-level assurance. CSC's [Sensitive Data](../data/sensitive-data/findata-permit.md) services also require higher levels of authentication.
 
-To raise the assurance level, you need to login to our customer portal MyCSC and download the Candour ID app. 
+To raise the assurance level to high, you need to login to our customer portal MyCSC and download the Candour ID app. 
 
 Please follow our instructions precisely. The identity vetting service provider is [Candour Ltd.](https://candour.fi/)
 

mkdocs.yml

@@ -286,6 +286,7 @@ nav:
       - Applying for Billing Units: accounts/how-to-apply-for-billing-units.md
       - Increasing disk quotas: accounts/how-to-increase-disk-quotas.md
       - Accessing Mahti large partition: accounts/how-to-access-mahti-large-partition.md
+      - Accessing Roihu large partition: accounts/how-to-access-roihu-large-partition.md
       - Reviewing Billing Unit usage: accounts/how-to-view-billing-unit-usage.md
       - Billing: accounts/billing.md
       - Multi-factor authentication: accounts/mfa.md