CERN Releases 300TB Of Large Hadron Collider Data

[Megalith]

CERN has begun releasing data dumps for all of you wannabe particle physicists or data hoarders out there.

CERN just dropped 300 terabytes of hot collider data on the world and you know you want to take a look. Kati Lassila-Perini, a physicist who works on the Compact Muon Solenoid (!) detector, gave a refreshingly straightforward explanation for this huge release. “Once we’ve exhausted our exploration of the data, we see no reason not to make them available publicly,” she said in a news release accompanying the data. “The benefits are numerous, from inspiring high school students to the training of the particle physicists of tomorrow. And personally, as CMS’s data preservation coordinator, this is a crucial part of ensuring the long-term availability of our research data.”

 

[CaptNumbNutz]

Knowledge can only benefit humanity. The only problem is having the bandwidth and the drive space to download 300TB.

 

[schizrade]

Gonna need a few more drive arrays... 300tb raw. Triple/quadruple that for any kind of processed data to analyze.

 

[tazeat]

300tb isn't really all that much anymore, will be significantly more when processed to a usable form likely. You can fit over a PB in a single rack these days. Still easily half million to a million (or more) for a vendor furnished solution.... Then you need a processing cluster.

 

[LightningCrash]

300tb isn't really all that much anymore, will be significantly more when processed to a usable form likely. You can fit over a PB in a single rack these days. Still easily half million to a million (or more) for a vendor furnished solution.... Then you need a processing cluster.

You can fit 1PB in 4U, or about 8PB per rack.

 

[GaryJohnson]

300tb isn't really all that much anymore, will be significantly more when processed to a usable form likely. You can fit over a PB in a single rack these days. Still easily half million to a million (or more) for a vendor furnished solution.... Then you need a processing cluster.

Or you could store 300TB on 20 LTO7 data tapes for about $2500 (not including the cost of the library and tape drives).

 

[westrock2000]

Wow, just looking through some of this data, on page 5.823E^15, if I am reading this right it looks like particle H367x20-0 publicly pairs with particle J90-2888d. However it looks like he might have an out of town entanglment with this other particle....who is the SAME type of particle. How has the main stream media not picked up this scandalous affair??

 

[defaultluser]

Read title as:

Large hardon collider data

Was expecting porn site link, was disappointed.

 

[Yakk]

Reminds me of something i used to say...

Access to information is good.

Understanding the information is better.

Access, understanding, and being able to work with the information is actually useful.

 

[tazeat]

Or you could store 300TB on 20 LTO7 data tapes for about $2500 (not including the cost of the library and tape drives).

Somehow I figure that you can't really perform big data computations on data from tape very effectively....

 

[Trimlock]

Would only take me 2087 days to download this at best case

 

[GaryJohnson]

Somehow I figure that you can't really perform big data computations on data from tape very effectively....

Well no, you'd linearly run through the full dataset on tape using some criteria to filter the records and create a reduced dataset on real hard drives somewhere that is small enough to actually analyze.

 

[LightningCrash]

Well no, you'd linearly run through the full dataset on tape using some criteria to filter the records and create a reduced dataset on real hard drives somewhere that is small enough to actually analyze.

You could tier it into something like SAM-FS so you can keep it mostly on tape.