Budgeting a production Galaxy instance
Hello all, We (at Tel Aviv University) are considering setting up a Galaxy instance (private, but production-geared). The capacity planning pages and surveys are great for hardware budgeting, but I want to ask about manpower. We’re looking forward to roughly 50 active users, and we’ll need to provide troubleshooting, training and integrating and writing new tools. In your experience, what is be the scope of a Galaxy admin’s position? Part-time? Full-time? Something else? Is more than one person needed to handle everything? Thank you, Vladimir Gritsenko
Hi Vladimir It is difficult to give you a good number, as it depends on a lot of factors, like: - how "active" are your anticipated '50 active user'? more than 1 job/workflow a day or just 1 job/workflow a month? - do you have to set up a system in order to give the galaxy users access to 'external' data, like fastq files? - how much training and what kind of training will they need - what kind of tools do you wanna integrate and/or write - do you have an IT-department which will help you, or do you have to do all the sysadmin stuff for the cluster as well To start, I would recommend at least two full time positions. This will give you enough manpower to set up the instance and provide training in the initial phase. And later, you will have enough flexibility to integrate new tools, follow the developments in the wider galaxy universe...and eventually give something back to the galaxy community. Regards, Hans-Rudolf On 10/05/2016 10:55 PM, Vladimir Gritsenko wrote:
Hello all,
We (at Tel Aviv University) are considering setting up a Galaxy instance (private, but production-geared). The capacity planning pages and surveys are great for hardware budgeting, but I want to ask about manpower. We’re looking forward to roughly 50 active users, and we’ll need to provide troubleshooting, training and integrating and writing new tools. In your experience, what is be the scope of a Galaxy admin’s position? Part-time? Full-time? Something else? Is more than one person needed to handle everything?
Thank you, Vladimir Gritsenko ___________________________________________________________ Please keep all replies on the list by using "reply all" in your mail client. To manage your subscriptions to this and other Galaxy lists, please use the interface at: https://lists.galaxyproject.org/
To search Galaxy mailing lists use the unified search at: http://galaxyproject.org/search/mailinglists/
Hello Hans-Rudolf, Thank you for your input! 1) By active users I'm thinking closer to one workflow per day (as an average, in reality I expect the workflows to come in batches spread out across the month). But we do have scalable IT infrastructure here, so that should help with changing loads. 2) Not sure what you mean by "external" data - they should definitely have the option to upload their own datasets, but I imagine that's the default option? If you mean connecting Galaxy to existing dataset repositories (e.g., sequencing centers) - I'm still exploring this need. 3) Regarding training, the baseline should be a workshop every semester (for newly-arrived graduate students), which is not that much, now that I think about it. But again, I wonder what other universities have implemented in this regard. 4) Regarding tools, good question! My own lab does NGS (mostly SNP calling and analysis), and I got to do some de novo assembly, but the University is large(-ish?) and ultimately I expect a diverse set of requirements from the various labs. But even for our own narrow purposes, we'd like the updated GATK tools, better SnpEff integration, etc. So I do expect a relatively high demand for tool integration. 5) We do have an IT department, so they'll handle running the hardware itself. We may also want to recruit Galaxy admins from the IT ranks. Regarding two full time positions - is this the standard (or average) in Galaxy set ups? I'm mostly concerned that after the initial set up and user growth stages there won't be enough work left.
On 6 Oct 2016, at 10:04, Hans-Rudolf Hotz <hrh@fmi.ch> wrote:
Hi Vladimir
It is difficult to give you a good number, as it depends on a lot of factors, like:
- how "active" are your anticipated '50 active user'? more than 1 job/workflow a day or just 1 job/workflow a month?
- do you have to set up a system in order to give the galaxy users access to 'external' data, like fastq files?
- how much training and what kind of training will they need
- what kind of tools do you wanna integrate and/or write
- do you have an IT-department which will help you, or do you have to do all the sysadmin stuff for the cluster as well
To start, I would recommend at least two full time positions. This will give you enough manpower to set up the instance and provide training in the initial phase. And later, you will have enough flexibility to integrate new tools, follow the developments in the wider galaxy universe...and eventually give something back to the galaxy community.
Regards, Hans-Rudolf
On 10/05/2016 10:55 PM, Vladimir Gritsenko wrote:
Hello all,
We (at Tel Aviv University) are considering setting up a Galaxy instance (private, but production-geared). The capacity planning pages and surveys are great for hardware budgeting, but I want to ask about manpower. We’re looking forward to roughly 50 active users, and we’ll need to provide troubleshooting, training and integrating and writing new tools. In your experience, what is be the scope of a Galaxy admin’s position? Part-time? Full-time? Something else? Is more than one person needed to handle everything?
Thank you, Vladimir Gritsenko ___________________________________________________________ Please keep all replies on the list by using "reply all" in your mail client. To manage your subscriptions to this and other Galaxy lists, please use the interface at: https://lists.galaxyproject.org/
To search Galaxy mailing lists use the unified search at: http://galaxyproject.org/search/mailinglists/
participants (2)
-
Hans-Rudolf Hotz -
Vladimir Gritsenko