A few introductory words

[Rakhirandiseo]

Alexey Poletaev | 06.08.2018
IncreaseFig. 1. General architecture of GridGain
Fig. 1. General architecture of GridGain
IncreaseFig. 2. Data Grid
Fig. 2. Data Grid
IncreaseFig. 3. Compute Grid
Fig. 3. Compute Grid
IncreaseRice. 4. Data Grid + Compute Grid
Rice. 4. Data Grid + Compute Grid
IncreaseFig. 5. Service Grid
Fig. 5. Service Grid
IncreaseFig. 6. Ignite streaming
Fig. 6. Ignite streaming
IncreaseFig. 7. Overall db-engines rating for August 2018.
Fig. 7. Overall db-engines rating for August 2018.
IncreaseFig. 8. db-engines rating for Relational DBMS for August 2018.
Fig. 8. db-engines rating for Relational DBMS for August 2018.
IncreaseFig. 9. db-engines ranking for Key-value Stores for August 2018.
Fig. 9. db-engines ranking for Key-value Stores for August 2018.
IncreaseFig. 10. Search history for GridGain
Fig. 10. Search history for GridGain
Unfortunately, there are very few technical materials on the topic. The Internet mainly contains videos with presentations on the YouTube channel from developers. Attempts to find at least some detailed technical description of the GridGain architecture or real technical implementation do not bring results. The Internet is overflowing with advertising, general presentations, links to the manufacturer's website and articles about Sberbank. Not a single live case of migration to macedonia whatsapp data with an analysis of the pros and cons, problem solving. Specialists try to discuss some near-technical problems on specialized forums, but there is very little information there. It is very strange to see this for a product with a history of more than 11 years! It seems to me that too much noise has been created around GridGain and good marketing has been carried out.

The technology description below is taken from Apache Ingnite, but it is also valid for GridGain.

What is GridGain?
Apache Ingnite (GridGain) is, in general terms, an open source platform for distributed storage and computation of large amounts of data across a cluster of nodes.

Apache Ignite Database uses RAM as the default storage and processing layer, so it is classified as an in-memory computing platform. The disk layer is optional, but once enabled, the disks will store the full data set, while the memory layer will cache the full or partial data set depending on its capacity. Apache Ignite can use RDBMS, NoSQL, or Hadoop databases as the disk layer.

Regardless of the API used, data in Ignite is stored as key-value pairs. The database component scales horizontally by distributing key-value pairs across the cluster so that each node owns a portion of the overall data set. Data is automatically rebalanced whenever a node is added to or removed from the cluster.