Feeds

Kaminario array slips on protection but won't go down on you

K2 flash box now highly available for enterprises

Internet Security Threat Report 2014

Kaminario has added enterprise disk drive array-like protection and availability features to K2, its solid state SAN storage boxes. You won't need to take one of those machines down, ever.

These are built from a grid or cluster of data nodes, blade servers, the K2-D DRAM array, K2-F flash array with Fusion-io ioDrives, and K2-H hybrid DRAM-flash products. The systems scale by adding more data nodes.

Kaminario says this set of availability and protection features is unique and can't be obtained from other flash array vendors, not that you'll lose much data if a flash array's power supply goes phut - but a DRAM array or a hybrid? That's a different and very volatile story.

CEO Dani Golan argues: "Customers need the same level of protection and operational efficiency as legacy storage. Without it solid-state storage can't penetrate the enterprise market."

It is adding N+1 hardware redundancy, automatic failover and non-disruptive operation. It should be fast; there are no spinning disks to slow down failover. The company is also adding something called RAID 10HD, which is mirroring and striping, high-volume snapshots - you can have more than 8,000 of them - and asynchronous replication, both host-based and array-based, so two K2s can keep in step.

The striping and mirroring is over different data nodes and media, providing protection against node failures. In the event of a failed data node, reads and writes use the mirrored nodes with the spare node being brought into play. A demo video shows a K2 data node being pulled out of the array and the system just keeps on working; there's a momentary blip, a couple of seconds, while it reconfigures I/O and presses the spare node into operation, and then just carries on.

Golan said: "In 10 seconds we're back to 85 per cent of the original performance. It sometimes takes hours to recover from a failure like that in legacy storage."

Kaminario recommends using its K2-D, with up to 25TB of DRAM, for write-intensive and latency-sensitive high-end OLTP and DBMS applications. The availability and protection features will reassure customers neither data nor online availability will be lost to hardware or power failures.

The K2-H is for mixed read and write DBMS and analytics situations, with DRAM used for writes and flash for reads. The availability and protection features add reassurance for the DRAM write data as well as for device availability. Ditto for the K2-F which is for read-intensive workloads - DBMS and analytics again - but doesn't have such a requirement for fast writes and won't lose information if there is a flash crash, except for the most recent data that might not have been committed to the flash.

Kaminario is also adding better performance monitoring software.

We asked about deduplication in K2 and there isn't any, Kaminaro prioritising latency, with Golan saying: "The world of storage is becoming more and more random. For that latency is the most important feature. We do 120 microseconds. Deduplication hurts latency significantly. We're very proud of our latency."

The availability and protection features are all up for grabs now, except for array-based replication which is coming in the second half of the year. K2 pricing with the data protection features starts at $20/GB with the K2-F. ®

Beginner's guide to SSL certificates

More from The Register

next story
NSA SOURCE CODE LEAK: Information slurp tools to appear online
Now you can run your own intelligence agency
Azure TITSUP caused by INFINITE LOOP
Fat fingered geo-block kept Aussies in the dark
Yahoo! blames! MONSTER! email! OUTAGE! on! CUT! CABLE! bungle!
Weekend woe for BT as telco struggles to restore service
Cloud unicorns are extinct so DiData cloud mess was YOUR fault
Applications need to be built to handle TITSUP incidents
Stop the IoT revolution! We need to figure out packet sizes first
Researchers test 802.15.4 and find we know nuh-think! about large scale sensor network ops
Turnbull should spare us all airline-magazine-grade cloud hype
Box-hugger is not a dirty word, Minister. Box-huggers make the cloud WORK
SanDisk vows: We'll have a 16TB SSD WHOPPER by 2016
Flash WORM has a serious use for archived photos and videos
Astro-boffins start opening universe simulation data
Got a supercomputer? Want to simulate a universe? Here you go
Do you spend ages wasting time because of a bulging rack?
No more cloud-latency tea breaks for you, users! Get a load of THIS
prev story

Whitepapers

Driving business with continuous operational intelligence
Introducing an innovative approach offered by ExtraHop for producing continuous operational intelligence.
A strategic approach to identity relationship management
ForgeRock commissioned Forrester to evaluate companies’ IAM practices and requirements when it comes to customer-facing scenarios versus employee-facing ones.
How to determine if cloud backup is right for your servers
Two key factors, technical feasibility and TCO economics, that backup and IT operations managers should consider when assessing cloud backup.
High Performance for All
While HPC is not new, it has traditionally been seen as a specialist area – is it now geared up to meet more mainstream requirements?
Internet Security Threat Report 2014
An overview and analysis of the year in global threat activity: identify, analyze, and provide commentary on emerging trends in the dynamic threat landscape.