Tutorial for using Allas in the Roihu supercomputer

One of the main use cases of Allas is to store data while it is not actively used on CSC supercomputers. When you start working, you stage in the data from Allas. And when the data is no longer actively used, it can be staged out to Allas. The scratch disk area of Roihu allows you to analyze large datasets. However, scratch disk is not intended for long-term storage of research data. Data that needs to be stored for a longer time than just a few weeks should be copied to Allas or LUMI-O object storage service instead. Allas provides a platform that you can use to store your data as long as your CSC project is active. In addition to storage, Allas can be used for transporting data between different servers and sharing data with other users.

If you have not used Allas before, then start by reading Allas general introduction, which includes many important things to know.

Note that CSC computing projects do not have access to Allas automatically. The project manager has to activate the Allas service in MyCSC before the project can start using Allas. Allas introduction page describes how to do it and also about default quotas and how to apply for more storage space.

Additional information regarding:

There are several alternative tools to move data between Allas and Roihu. You can use the Roihu web interface, commandline tools or Python, R or other tools. In Roihu, the Allas commandline tools are installed by CSC and provided through the allas module.

Allas module in Roihu

Examples

This tutorial provides three examples for using Allas in Roihu. The examples are based on interactively executed commands, and thus examples 1 and 2 apply only for relatively small datasets (max. some hundreds of GBs). The third example suits also for bigger datasets.

The first example uses the a-commands (a-put, a-get) for uploading data from Roihu to Allas, and then downloading the data back to Roihu.
The second example transfers the same data using Rclone.
The third example handles the case in which the dataset to be copied includes a large amount of files.

The a-commands are better suited for cases where the data is mainly used within the CSC computing environment (Roihu, Lumi). The second option, Rclone, is good for cases when the data will be used outside CSC too. Note that s3cmd and aws s3 are totally valid alternatives for rclone and a-commands, even though they are not used in this tutorial.

Example 1: Using Allas with a-commands

A. Uploading data from Roihu to Allas

The a-commands are Allas-specific tools that allow an easy start with Allas. The a-commands archive and move data automatically. You can also compress your data before storage. For example, for text-formatted data compression reduces the storage space needed, but on the other hand makes the transfer process slightly slower. The a-commands are a good option for miscellaneous data that is mostly used in the CSC environment.

In this example, we have a subdirectory genomes/zebrafish in the scratch directory of a project in Roihu (/scratch/project_2001659). The zebrafish directory contains eight files listed below:

[kkayttaj@roihu-cpu-login1 ~]$ ls /scratch/project_2001659/genomes/zebrafish
Danio_rerio.GRCz10.91.1.bt2      Danio_rerio.GRCz10.91.2.bt2  
Danio_rerio.GRCz10.91.3.bt2      Danio_rerio.GRCz10.91.4.bt2  
Danio_rerio.GRCz10.91.rev.1.bt2  Danio_rerio.GRCz10.91.rev.2.bt2  
Danio_rerio.GRCz10.fa            Danio_rerio.GRCz10.fa.fai

To copy the content of this directory to Allas, we first set up the Allas environment:

module load allas

Then, we open a connection to Allas using the command allas-conf. The command asks for the user's CSC password and then lists the Allas projects that are accessible. In this case, we select project_2001659.

[kkayttaj@roihu-cpu-login1 ~]$ allas-conf
Mode s3cmd
Please enter CSC password for account kkayttaj: <password>
Checking projects available for your account.
Please wait.
1) project_2000982     2) project_2001659     3) project_2000136      4) abort allas_conf
Please choose a project by giving an item number from the list above: 2
Configuration will be done for project: project_2001659
Protocols:
  S3
s3cmd configuration updated

Remote endpoints configured for rclone

Rclone remote name       Protocol
--------------------------------------------------------
s3allas:                 s3    
s3allas-project_2001659: s3    
--------------------------------------------------------
rclone remote s3allas: now provides an S3 based connection to project project_2001659 in Allas
aws client and s3cmd use now project_2001659
a-commands use now project_2001659 when used in S3 mode

allas-conf stores S3 authentication keys to your home directory. If we want to start using another project, we need to run allas-conf again.

Next, we enter the zebrafish directory:

cd /scratch/project_2001659/genomes/zebrafish

We can now upload files one by one to Allas using the a-put command:

a-put Danio_rerio.GRCz10.fa

At the end of the upload process, the command reports:

-------------------------------------------------------------------------------
1 files from Danio_rerio.GRCz10.fa uploaded to bucket 2001659-mahti-SCRATCH in Allas as one file: 
2001659-mahti-SCRATCH/genomes/zebrafish/Danio_rerio.GRCz10.fa
-----------------------------------------------------------------

Upload summary:
              Date                      Name  Files Size(kB)         Location in allas
 12.10.20 12:10:50     Danio_rerio.GRCz10.fa      1  1330852 2001659-mahti-SCRATCH/genomes/zebrafish
-----------------------------------------------------------------
OK

Moving data to Allas file by file is slow and produces a large amount of objects. It is often more efficient to upload data to Allas one directory at a time and store the data in bigger chunks. For example, to upload the zebrafish directory, we first enter the parent directory genomes:

cd /scratch/project_2001659/genomes/

Then, we use a-put to upload the whole zebrafish directory to Allas as a single object:

a-put zebrafish/

At the end of the upload process, the command reports:

Files or directories to be uploaded: zebrafish/
Processing: zebrafish/
Checking total size of zebrafish. Please wait.

zebrafish contains 8 files or directories that take 2585920K of disk space
Collecting data from directory zebrafish to archive file: zebrafish.tar
zebrafish/
zebrafish/Danio_rerio.GRCz10.fa.fai
zebrafish/Danio_rerio.GRCz10.91.3.bt2
zebrafish/Danio_rerio.GRCz10.91.4.bt2
zebrafish/Danio_rerio.GRCz10.fa
zebrafish/Danio_rerio.GRCz10.91.rev.2.bt2
zebrafish/Danio_rerio.GRCz10.91.rev.1.bt2
zebrafish/Danio_rerio.GRCz10.91.2.bt2
zebrafish/Danio_rerio.GRCz10.91.1.bt2
Packing successful.
Uploading data to s3allas.
Transferred:        3.044 GiB / 3.044 GiB, 100%, 62.753 MiB/s, ETA 0s
Transferred:            1 / 1, 100%
Elapsed time:        46.9s
Confirming upload...

Adding metadata for uploaded zebrafish
Transferred:            975 B / 975 B, 100%, 0 B/s, ETA -
Transferred:            1 / 1, 100%
Elapsed time:         0.0s

-------------------------------------------------------------------------------
8 files from zebrafish uploaded to bucket 2001659-roihu-scratch in Allas as one tar file: 
2001659-roihu-scratch/genomes/zebrafish.tar
-----------------------------------------------------------------

Upload summary:
              Date                      Name  Files Size(kB)         Location in Allas
 06.05.26 13:05:05                 zebrafish      8  2585920 2001659-roihu-scratch/genomes
-----------------------------------------------------------------
OK

After this, we have another object in the 2001659-roihu-scratch bucket:

[kkayttaj@roihu-cpu-login1 genomes]$ a-list 2001659-roihu-scratch
2001659-roihu-scratch/genomes/zebrafish.tar
2001659-roihu-scratch/genomes/zebrafish/Danio_rerio.GRCz10.fa

Note that the file Danio_rerio.GRCz10.fa is in fact now stored in Allas twice; both as an individual object (genomes/zebrafish/Danio_rerio.GRCz10.fa) and as part of the genomes/zebrafish.tar object.

B. Downloading data to Roihu

Next, we download the same data back to Roihu. In this example, we assume that zebrafish genome data has been removed from Roihu, either by the user or because of the automatic cleaning process.

After connecting to Roihu, we go to the scratch directory of project 2001659 and load the allas module:

cd /scratch/project_2001659
module load allas

From the output of the module load allas command you can check of the a-commands use the right Allas, in this case project_2001659. If needed you can take the project in use with allas-conf command:

allas-conf project_2001659

In the case of the command above the configuration process asks only for the CSC password and then sets up the connection to Allas for project 2001659. As the Roihu scratch directory is shared by all project members, we create a user-specific subdirectory kkayttaj:

mkdir kkayttaj
cd kkayttaj/

With the command a-list, we can now see the objects that were previously uploaded to Allas:

[kkayttaj@roihu-cpu-login2 kkayttaj]$ a-list
2001659-roihu-scratch
[kkayttaj@roihu-login1 kkayttaj]$ a-list 2001659-roihu-scratch
2001659-roihu-scratch/genomes/zebrafish.tar
2001659-roihu-scratch/genomes/zebrafish/Danio_rerio.GRCz10.fa

Locating data is easy as there are only two objects in the bucket, but as more data is added to Allas, locating a specific file among dozens of buckets containing hundreds of objects may be difficult. In that case, you can search for a specific file using the command a-find. In this example, we can check if an object contains the file Danio_rerio.GRCz10.fa:

[kkayttaj@roihu-cpu-login2 kkayttaj]$ a-find -a Danio_rerio.GRCz10.fa
Checking bucket: 2001659-roihu-scratch

Matching object names:
2001659-roihu-scratch/genomes/zebrafish/Danio_rerio.GRCz10.fa
----------------------------------------------
a-put generated objects that include matching file name:

 Object: 2001659-roihu-scratch/genomes/zebrafish.tar
 includes 2 file names that that match query: Danio_rerio.GRCz10.fa

 Object: 2001659-roihu-scratch/genomes/zebrafish/Danio_rerio.GRCz10.fa
 includes 1 file names that that match query: Danio_rerio.GRCz10.fa

------------------------------------------------ 
Query: Danio_rerio.GRCz10.fa
Total of 3 hits were found in 2 objects
-------------------------------------------------

The a-find report above shows, for example, that the object 2001659-mahti-SCRATCH/genomes/zebrafish.tar contains two files whose names match Danio_rerio.GRCz10.fa (the other file is Danio_rerio.GRCz10.fa.fai). Note that a-find finds matches only among objects that were uploaded with a-put. Further, a-find searches objects with object name only. Don't include the bucket name to the search term. Instead use option -b <bucket-name> if you wish to search the object from just one bucket.

Next, we download the data to Roihu using the a-get command:

[kkayttaj@roihu-cpu-login2 kkayttaj]$ a-get 2001659-roihu-scratch/genomes/zebrafish.tar
Starting to copy data from allas...
Object:
  2001659-roihu-scratch/genomes/zebrafish.tar 
copied and uncompressed from s3allas into:
  zebrafish

After this, the current working directory in Roihu has a new directory zebrafish that contains the files that were previously uploaded from Roihu to Allas:

[kkayttaj@roihu-cpu-login2 kkayttaj]$ ls zebrafish/
Danio_rerio.GRCz10.91.1.bt2  Danio_rerio.GRCz10.91.rev.1.bt2
Danio_rerio.GRCz10.91.2.bt2  Danio_rerio.GRCz10.91.rev.2.bt2
Danio_rerio.GRCz10.91.3.bt2  Danio_rerio.GRCz10.fa
Danio_rerio.GRCz10.91.4.bt2  Danio_rerio.GRCz10.fa.fai

Example 2: Using Allas with Rclone

A. Uploading data with Rclone

Rclone is the power-user tool for Allas. It is good in cases where the data must be stored such that each file is a separate object.

Warning

Rclone provides a fast and efficient way to use Allas, but you should use it carefully as Rclone operations can overwrite and remove data both in Allas and in the local disk environment without notifying or asking for confirmation.

Using Allas with Rclone from Roihu

This example uses the same data as the previous case: in the scratch directory of Roihu, we have a subdirectory genomes/zebrafish that contains the eight files listed below:

[kkayttaj@roihu-cpu-login2 ~]$ ls /scratch/project_2001659/genomes/zebrafish
Danio_rerio.GRCz10.91.1.bt2  Danio_rerio.GRCz10.91.rev.1.bt2
Danio_rerio.GRCz10.91.2.bt2  Danio_rerio.GRCz10.91.rev.2.bt2
Danio_rerio.GRCz10.91.3.bt2  Danio_rerio.GRCz10.fa
Danio_rerio.GRCz10.91.4.bt2  Danio_rerio.GRCz10.fa.fai

To copy the content of this directory to Allas, we first set up the Allas environment:

module load allas

From the output of the module load allas command you can check whether the a-commands use the right Allas, in ths case project_2001659. If needed you can take the project in use with allas-conf command:

allas-conf project_2001659

In the case of the command above the configuration process asks only for the CSC password and then sets up the connection to Allas for project 2001659. After configuration, you can use the Allas area of the defined project using two rclone endpoints: s3allas:, s3allas-project_name. In this example, the endpoints are s3allas: and s3allas-project_2001659:.

After configuration, these two endpoints are in use in all Roihu sessions, including batch jobs. The short endpoint is nicer to use in command line, but it changes if you run allas-conf to use another project. Thus, the project-specific endpoint (s3allas-project_2001659) would be a better option for batch jobs and project-specific scripts.

To start the upload, we go to the genomes directory:

cd /scratch/project_2001659/genomes

Instead of a-put that was used in the previous example, we use command rclone copy to copy all files from the given directory to Allas. In the case of rclone, there is no default bucket. Instead, we have to define a bucket. In this example, we use the bucket name 2001659-genomes and define each object name to have the prefix zebrafish.

rclone copy -P zebrafish/ s3allas:2001659-genomes/zebrafish

After copying the files, we can use rclone ls to see what has been uploaded to Allas:

[kkayttaj@roihu-cpu-login2 genomes] rclone ls s3allas:2001659-genomes/zebrafish
450646234 Danio_rerio.GRCz10.91.1.bt2
334651392 Danio_rerio.GRCz10.91.2.bt2
   187325 Danio_rerio.GRCz10.91.3.bt2
334651387 Danio_rerio.GRCz10.91.4.bt2
450646234 Danio_rerio.GRCz10.91.rev.1.bt2
334651392 Danio_rerio.GRCz10.91.rev.2.bt2
1362788082 Danio_rerio.GRCz10.fa
      715 Danio_rerio.GRCz10.fa.fai

B. Downloading the data to Roihu

Next, we download the same data back to Roihu. After connecting to Roihu, we go to the scratch directory of project_2001659 and load the allas module:

cd /scratch/project_2001659
module load allas

If rclone is still configured for project 2001659, there is no need to run allas-conf.

As the Roihu scratch directory is shared by all project members, we create a user-specific subdirectory kkayttaj and go there:

mkdir kkayttaj
cd kkayttaj/

We can now use the command rclone lsd to check the available buckets in Allas:

[kkayttaj@roihu-cpu-login2 kkayttaj]$ rclone lsd s3allas:
  3268222761 2026-05-03 10:01:42         8 2001659-genomes
  2576778428 2026-05-03 10:01:42         4 2001659-roihu-SCRATCH

Now we see two buckets. 2001659-genomes is the one that was just created in this example, while 2001659-roihu-SCRATCH originates from the previous a-command example. Next, we list the objects in the 2001659-genomes bucket:

[kkayttaj@roihu-cpu-login2 kkayttaj]$ rclone ls s3allas:2001659-genomes
450646234 zebrafish/Danio_rerio.GRCz10.91.1.bt2
334651392 zebrafish/Danio_rerio.GRCz10.91.2.bt2
   187325 zebrafish/Danio_rerio.GRCz10.91.3.bt2
334651387 zebrafish/Danio_rerio.GRCz10.91.4.bt2
450646234 zebrafish/Danio_rerio.GRCz10.91.rev.1.bt2
334651392 zebrafish/Danio_rerio.GRCz10.91.rev.2.bt2
1362788082 zebrafish/Danio_rerio.GRCz10.fa
      715 zebrafish/Danio_rerio.GRCz10.fa.fa

Finally, we use the rclone copy command to copy the data from Allas to Roihu into a new directory zebrafish2:

[kkayttaj@roihu-cpu-login2 kkayttaj]$ rclone -P copy s3allas:2001659-genomes/zebrafish zebrafish2
Transferred:        3.044 GiB / 3.044 GiB, 100%, 323.600 MBytes/s, ETA 0s
Transferred:            8 / 8, 100%
Elapsed time:        9.6s

[kkayttaj@roihu-cpu-login1 kkayttaj]$ ls zebrafish2
Danio_rerio.GRCz10.91.1.bt2      Danio_rerio.GRCz10.91.3.bt2  
Danio_rerio.GRCz10.91.2.bt2      Danio_rerio.GRCz10.91.4.bt2
Danio_rerio.GRCz10.91.rev.1.bt2  Danio_rerio.GRCz10.fa
Danio_rerio.GRCz10.91.rev.2.bt2  Danio_rerio.GRCz10.fa.fai

Example 3: Uploading complex directory structures to Allas

Some workflows and software create complex directory structures to store and manage data. You might have directories that have thousands or even millions of individual files. Copying such datasets to Allas takes time and is not always straightforward. The most reasonable way to upload this kind of data depends on the case. This example introduces a few alternatives.

First, we open a tmux session on Roihu and set up an Allas connection just like in the previous example:

ssh <username>@roihu-cpu.csc.fi
tmux new -s allas-upload
module load allas
allas-conf

Suppose we have a directory structure that contains images from road condition cameras from ten locations with an interval of ten minutes from the years 2021–2025. The data is located in the directory road_cameras where each location has its own subdirectory (ten directories). Inside each subdirectory, there is another layer of subdirectories, one for each year (five subdirectories), each containing subdirectories for every day of the year (further 365 subdirectories), each containing 144 small image files.

For example:

road_cameras/site_7/2022/day211/image_654887.jpg

Thus, the total number of files in the road_cameras directory is 10 * 5 * 365 * 144 = 2 628 000.

In principle, we could copy all 2.6 million files as separate objects to Allas, but in that case, we should split the data into multiple buckets as one bucket can have at most 0.5 million objects. You could, for example, run a separate rclone command for each site_* directory and put the data from each site to a site-specific bucket. For example:

rclone copyto road_cameras/site_1 s3allas:2001659_road_cameras_site_1/

This way, you would end up creating ten buckets each containing 262 800 objects. This approach could be the most efficient way for storing and reusing the data if you know that you will need to access individual images randomly. However, you would then need to use some other tools than rclone or a-commands, as rclone has a bug that makes it fail in cases where one bucket contains a huge number of object names (typically this starts to affect in cases where the number of objects is over 15 000- 50 000).

As another extreme option, we could use a-put and collect all data into a single archive object. In order to do that, you must add the option --skip-filelist to the a-put command. By default, a-put collects detailed metadata of each file in an ameta file. However, if you have millions of files, collecting this information takes a long time. If you need to know the file names, you can use the --simple-fileslist option to collect the names – but no other information – of the files in the metadata file. This already speeds up the preprocessing significantly. However, as in this case the naming has been systematic, storing the file names to the metadata files can be just ignored altogether (--skip-filelist), which is the fastest option.

a-put --skip-filelist road_cameras/

This approach would store all 2,6 million files as a single object.

In practice, however, the optimal way of storing the data is often something between these two extremes. As a compromise, you could apply packing at a higher level in the hierarchy. For example:

a-put --skip-filelist road_cameras/site_*

This would produce ten objects, each containing all information from a single camera site. Alternatively, you could do the archiving so that data from each year from each camera is collected as a single object:

a-put --skip-filelist road_cameras/site_*/20*

This option would store the data as 50 objects. Day-based objects for each camera might be the most practical option for using the data later on, but as a downside, preprocessing the data into 10 * 5 * 365 = 18250 objects probably takes quite a long time.

Copying millions of files to Allas takes a long time regardless of the method. If we have started the a-put command inside a tmux session, we can detach from the virtual session by pressing Ctrl-b d, log out from Roihu and leave the upload process running for long periods. However, if the login node reboots between the sessions, your tmux session will end.

To return to the session later, reconnect to Roihu and run:

tmux attach -t allas-upload

When you no longer need the tmux session, you can kill the session by typing exit inside the session.

Once the a-put command is finished, we will run a-check command to check if all the data objects have been created. a-check needs to be executed with the exact same options that were used with the a-put command. So in this case the command would be:

a-check --skip-filelist road_cameras/site_*/20*

The a-check command compares the item names to be uploaded to the matching objects in Allas. The files or directories that don't have a target object in Allas are reported and stored in a file. In this case, if some objects in the a-put command above would be missing, then a-check would list the missing files and directories in a file missing_<bucket_name>_<number> (the number at the end is just a random number).

The file of missing items can then be used with a-put option --input-list to continue the failed upload process:

a-put -b 2001659-uniref --nc --input-list missing_<bucket_name>_<number>

You should note that a-check does not check if the actual contents of the object are correct. It checks only the object names, which might as well originate from some other sources.