Home → CLC FAQ - Basics: Installation, data, system resources → System and resources → What is the difference between running a Genomics Server and running a Genomics Workbench on a large central machine that multiple users can access?
3.1. What is the difference between running a Genomics Server and running a Genomics Workbench on a large central machine that multiple users can access?
If you are working in a group with multiple users who wish to make use of the Genomics Workbench analysis tools on a central machine, it is possible for members of your group to
- share access to a single Genomics Workbench using a network license via remote desktop on the central machine, or
- connect to a Genomics Server running on the central machine from a local Genomics Workbench such as from the user's own laptop or a desktop computer.
such that multiple users can run analysis tasks on the large central machine. The primary differences between these two set ups are related to
- Job Queue
- Multiple Users
- Data Access and Permissions
The Genomics Server allows for a large number of jobs to be submitted to it such that each job submitted will be placed in a queue and executed when the available job slot is available. The Workbench, however, will not queue jobs, but will run them immediately, without regard for other concurrent processes or available resources. For example, if you have a network license for the Genomics Workbench, and five separate users choose to each runs a large de novo analysis on the central machine through remote desktop protocols such VNC or RDP. Each of those analyses thinks it has rights to the memory of the whole machine because the Genomics Workbench was designed to consider single user on a single machine. As a result, all jobs are launched as they are submitted. If simultaneously running analyses are large, it is very likely that the Genomics Workbench will crash with an out of memory error caused by resource competition.
Alternatively, if 5 de novo assembly jobs are submitted to the Genomics Server, they will be queued such that each will run one at a time or only one job will run per execution node. More information about the queue can be found
There are other benefits of the Genomics Server, especially if you are running a service. It is designed with the idea of multiple users, in a way that the Workbench is not. This includes the authentication system, the audit log, the ability to have installed Workflows that are available for all users, and the ability to control who has the rights to run certain tools or has the access to certain data folders. For example, maybe your service doesn't include de novo assembly. You can control access to allow the users to run all but the de novo assembly tools on the server.
Data Access and Permissions
The data access and permission configurations for the Genomics Server are described in
You can always configure the data file access and permissions (in .clc format) through filesystem or sharing security policy. They both overwrite the application level access control configurations. The application level configurations for data access can only be configured within the Genomics Server. The Workbenches, as described above, has single user design and does not have the in-program data file access and permission settings. If you have a centralized data sharing servers, then having the Genomics Server will be advantageous in terms of managing the CLC data through the application level settings.
There are also more general features available via a Genomics Server that are not available via the Workbench, such as the use of the Command Line Tools as a client (instead of the Genomics Workbench) or the ability to tie into other local tools or in-house pipelines using External Applications.
Of course, only you can decide what would suit you best, but unless you plan to offer your service to only a very small group of people who are in tight communication with each other (e.g. they will not all launch large jobs at the same time). Perhaps one way to consider this further is to work your way through the options in the administrative interface of the Genomics Server. Apart from the ability to set up File system locations (known as Data Locations in the Workbench), install Plugins, install Workflows and set up blast data locations, all the other functionalities referred to in that interface are only available on the Genomics Server. So this may help in your considerations.