CumuloNimbo Public Workshop at 2nd EUDAT Conference, Rome, Italy, 28th Oct. 2013

The CumuloNimbo has organized its public workshop at the 2nd EUDAT Conference in Rome (Italy) during 28th Oct. 2013.

The target of the conference was to create awareness within the science big data community.

The title of the workshop was: CumuloNimbo: The Sky is No Longer the Limit for Big Data!

The agenda of the workshop was:

- CumuloNimbo Overview.
- Ultra scalable transactional processing
- SQL support over NoSQL
- Beyond HBase
- CoherentPaaS and LeanBigData
- Discussion

The workshop journalistic description was:

Big Data: Only for analytical processing? These days that idea is ancient history. Now, with CumuloNimbo, Big Data can deal with any workload!
So far, Big Data has been a synonym for performing large analytical queries on massive amounts of data. However, in the CumuloNimbo workshop at the 2nd EUDAT conference, we will see how Big Data is getting even bigger, thanks to the new advances brought about by the CumuloNimbo project in ultra-scalable transactional processing.
Big Data is no longer constrained to just the analytical processing world. Unlike today, transactional workloads in the near future will scale out linearly without having to resort to sharding (which frequently requires that substantial modifications are made to applications). An extra benefit is that users will not have to resort to NoSQL (which currently implies sharding) in order to scale. If SQL is needed, it can be used and will scale out. If NoSQL is enough, it can be used and fewer resources will be consumed for query processing. If both SQL and NoSQL are needed, the two interfaces will be available to access the same data store in a coherent manner. Additionally, since updates will now scale as much as is needed, it will be possible to have Big Data combined with complex event processing (CEP) on massive event streams. This approach is based on enabling CEP queries to correlate these massive event streams over stored Big Data at network rates.
With these new advances, the Big Data world will be transformed to incorporate any workload from online analytical processing (OLAP) to online transaction processing (OLTP) and even to complex event processing. The new Big Data milieu will also support all the different data management paradigms available – from NoSQL technologies to CEP and SQL. Additionally, it will be possible to integrate specialized data stores (such as graph databases or document-oriented data stores) with SQL and thus applications will be able to exploit the best of the SQL and NoSQL worlds.