Great, thanks Tom. I’ll take a look at this.

 

Cheers,

Jez

 

 

 

There is a free Globus training course at Rutherford Appleton Laboratory on 10^th Jan

https://www.scd.stfc.ac.uk/Pages/The-Globus-World-Tour.aspx

 

 

Do you need to transfer and share huge amounts of research data? If you do, you might be interested in this training course.

 

The Scientific Computing Department (SCD), in collaboration with Globus Online, is hosting and managing a one-day training course on the Globus research data management service. 

 Globus was developed by the University of Chicago and is used by hundreds of thousands of researchers at institutions worldwide. In this training course, which is free of charge, the Globus team will present the inner workings of Globus and the means for using Globus as part of your research.

Who is it for?

This course will be useful if you are anticipating large amount of data being moved between data sources and your machine learning frameworks, and is targeted at:

• System administrators who are planning to deploy or use Globus at their institution;
• Researchers building applications and frameworks ;
• Anyone who is interested in learning more about the service for research data management.

What will you get out of it?

• Learn how the Globus platform simplifies development of applications for researchers;
• experiment with new Globus services and APIs;
• exchange ideas with peers on ways to apply Globus technologies; 
• expand your knowledge of Globus administration features.

​View the program

The training is being held at the STFC Rutherford Appleton Laboratory, Didcot, Oxfordshire, on 10th January 2019 but places are limited so REGISTER NOW to secure your place.

 

Best Regards,

Tom

 

 

Tom Griffin

Director, Scientific Computing Department

Science and Technology Facilities Council

 

[log in to unmask]

+44 (0)1235 445305

 

 

 

 

Hi folks, we’re currently looking at ways of improving the way we deliver the largest datasets on https://data.bl.uk/ to users. The largest today are several hundred GB, and it won’t be long before we’re into the TB. Large downloads can be a pain for users because they take a long time and can easily be interrupted. They also potentially present a significant cost to us as a data provider because outbound bandwidth costs can be high for data stored in the cloud. I know this is something that many people in the community will already have grappled with so I’m hoping there will be some experience to share.

 

Possibilities we’ve discussed so far include:

·         Just let people download over HTTP but advise use of a download manager to handle interruptions to the connection

·         Publish via BitTorrent, which  has the advantage that if a large number of people are downloading the same thing at once (e.g. during a workshop) our outgoing bandwidth use could be significantly less than filesize × number of people

·         Allow people to request a copy on disk via courier, probably charged to cover costs

·         Split datasets into smaller chunks to make it easier to get just the bit you need (but makes it more effort if you do want the whole lot)

·         Allow users to move their compute to the data, either in the cloud or by renting out space in our machine room (this is essentially what AWS have done for some big public datasets)

·         Provide a dedicated API and/or UI to allow users to browse the collection and select a custom subset to download

 

The last two would probably be my preference in the long run, but require the most resource to set up and maintain.

 

There was also some discussion a few years ago of using GridFTP but I don’t know where those went.

 

Any advice would be most welcome and I’d be happy to summarise responses for the community. Feel free to reply to me directly, and I’ll withhold or anonymise your advice from the summary on request.

 

Many thanks,

Jez

 

		Jez Cope (he/him) Data Services Lead Research Services	The British Library Building 6a Boston Spa Wetherby West Yorkshire LS23 7BQ www.bl.uk
		01937 546241 [log in to unmask]

 

 

To unsubscribe from the RESEARCH-DATAMAN list, click the following link:
https://www.jiscmail.ac.uk/cgi-bin/webadmin?SUBED1=RESEARCH-DATAMAN&A=1

 

To unsubscribe from the RESEARCH-DATAMAN list, click the following link:
https://www.jiscmail.ac.uk/cgi-bin/webadmin?SUBED1=RESEARCH-DATAMAN&A=1