Swiss financial services provider PostFinance has deployed 525TB of Cloudian intent storage in a ubiquitous pierce divided from record entrance storage as good as to support archiving for Hadoop and Apache Spark large information workloads.
PostFinance provides financial services to some-more than 3 million customers, has around 3,500 staff and operates dual datacentres.
Migrations towards intent storage are holding place for Hadoop and Apache Spark large information archives, as good as a pierce divided from record storage towards object storage.
Key reasons for this are scalability not found in file and retard entrance storage, and that S3-based intent storage brings ease-of-use for focus developers, pronounced comparison systems operative Andre Kocher.
PostFinance uses NetApp for record storage and Hitachi HDS storage for block. The latter is used for pivotal databases and there are no skeleton to quit information off this hardware.
However, there are skeleton to pierce 100 million files over to Cloudian intent storage, and that pierce will mostly impact existent NetApp hardware during PostFinance. So, because a change to intent storage?
“We’re always looking for what’s going to be in a future,” pronounced Kocher. “And intent storage is one of a technologies we identified.”
A pivotal motorist for PostFinance is that it is really elementary to use for focus developers. “With intent storage, we can give lots of shortcoming to focus developers. They can emanate their possess buckets as they work. Before they would have to ask record shares or LUNs in storage.
PostFinance has deployed 12 Cloudian HyperStore intent storage nodes with a sum of 525TB of ability – inside are SAS drives, and connectivity uses 10Gbps Ethernet. Cloudian’s HyperFile provides entrance to record data.
Cloudian’s core product is intent storage based on a Apache Cassandra open source distributed database. It can come as storage program to be deployed on commodity hardware, in cloud instances or in hardware appliance form. Its Hyperfile record entrance – that is Posix/Windows-compliant – can also be deployed on-premise and in a cloud to yield record access.
At PostFinance, emigration of record information is an ongoing process, while large information archiving is approaching to go into prolongation in Apr or May.
Another pivotal advantage for PostFinance is that S3 storage works good with containers, pronounced Kocher. “It’s easier to have an S3 endpoint than to have to map to determined storage.”
Is Kocher means to put any numbers on a advantages gained by a pierce to Cloudian? “No,” he said. “I can’t put anything into figures, though a large advantage is that a in-built information insurance in Cloudian means we don’t need backup for a platform.”
PostFinance’s Cloudian deployment uses 8+2 erasure coding, that sees information separate into 6 fragments with dual sets of relation data.
Article source: https://www.computerweekly.com/news/252456979/Swiss-bank-moves-away-from-file-with-Cloudian-object-storage
