On Day-2 SFD-7, we had a presentation from Springpath – a startup in Software defined Data platform. With the HALO (Hardware Agnostic Log-structure Objects) architecture and multi-platform support, Springpath Data Service (SDS) platform is soon going to be a big competitor to any of the market leaders in near future.
The technical discussion was lead by Mallik Mahalingam and Krishna Yadappanavar the founders of the company. As they know their product very much, they were able to get into detail about all the features, architecture and implementation. In this post here, let us discuss about the features of SDS from Springpath.
The Cluster : Springpath offers a hyper-convergence software solution – SDS – which can be run on various platforms including VMware vSphere, Hyper-V, KVM, Docker Containers etc… This multi-platform support is a great advantage for Springpath Data platform compared with the competitors. Springpath gives a software-only solution which can run on hardwares from majority of the vendors. SDS cluster makes use of the HDD and Flash drives from the hardware for datastores and the cache layer.
Springpath Data Platform is very simple and easy to manage as it is leveraging existing management tools and being basically a software-only solution. It can run on a minimum of three node cluster. And the system allows independent scaling of HDD nodes, SSD nodes (contributes towards the caching layer) or the compute nodes. The data gets automatically re-balanced across the tiers as new nodes are getting added to the cluster.
Caching : The caching system is really complex with the SDS platform and it plays an integral role in making the system work as cool as it is. The cache leveraged architecture reduces hotspots on a specific node and allows the content to be addressed across multiple nodes. The caching tier contains cache vNodes over the physical SSD nodes. The data striping is done internal to the VM layer (underlying VMDK layer) and is being distributed across the nodes. All the incoming writes are mirrored at the caching tier.
Features ++ : Multi-platform support, quick installation and provisioning, linear scale-out capability, easy addition/removal of nodes etc… are some of the big positives of SDS platform. Easier snapshot handling (Faster creation and deletion of pointer-based snapshots), faster inline dedupe and compression capabilities, advanced caching system which reduce the hotspot nodes, easy and used-to management with the use of existing management tools etc… also deserve a note.
Additional links :
Here you can find more posts from me on SFD7. Comment your thoughts…
Disclaimer: My travel and accommodation expenses for the SFD7 (Storage Field Day 7) event were paid by the Tech Field Day team. I am not requested to post about event content and I am not getting paid for any posts or for the time I spend during the event.