[Preamble: I have been invited by GestaltIT as a delegate to their Tech Field Day for Storage Field Day 18 from Feb 27-Mar 1, 2019 in the Silicon Valley USA. My expenses, travel and accommodation were covered by GestaltIT, the organizer and I was not obligated to blog or promote their technologies presented at this event. The content of this blog is of my own opinions and views]
Storage technology is complex. Storage infrastructure and data management operations are not trivial, despite what the hyperscalers like Amazon Web Services and Microsoft Azure would like you to think. As the adoption of cloud infrastructure services grow, the small and medium businesses/enterprises (SMB/SME) are usually left to their own devices to manage the virtual storage infrastructure. Cloud Service Providers (CSPs) addressing the SMB/SME market are looking for easier, worry-free, software-defined storage to elevate their value to their customers.
Managed high performance block storage
StorPool is a scale-out block storage technology, capable of delivering 1 million+ IOPS with sub-milliseconds response times. As described by fellow delegate, Ray Lucchesi in his recent blog, they were able to achieve these impressive performance numbers in their demo, without the high throughput RDMA network or the storage class memory of Intel Optane.
StorPool’s secret sauce and cluster architecture
Their secret sauce to achieve such high performance is their StorPool protocol. In the architecture diagram below,
I have highlighted that StorPool Block Device Driver on the client nodes in the diagram. The client nodes are Linux-based with XFS, ext4, ocfs2, GFS2 filesystems supported. The unique block device driver is installed as /dev/storpool/* in Linux, and presents itself as block device to file systems and applications north bound.
Each Linux client (and supported Linux-based virtualization technology – KVM, Docker/LXC, OpenVZ and VirtualBox), access the shared storage pool in the StorPool cluster in parallel, interacting with the StorPool Server Daemon in each server node in the cluster. This is an important point to note, because this is what gives StorPool the performance edge over other software-defined storage platforms.
This approach is very similar to several HPC (high performance computing) technology such as Lustre and BeeGFS, where the client-side driver controls the I/O to the server nodes in the cluster.
StorPool runs on standard x86 hardware with SSDs and HDDs, and each server node runs Linux. As more performance and capacity grow, more server nodes can be added to scale. The network can be standard Ethernet, RDMA and Infiniband.
It also presents block storage volumes to VMware ESXi and Hyper-V via the iSCSI protocol.
What I liked
What I liked at that Storage Field Day 18 session was how StorPool positioned itself in the market. It was not gunning for the enterprise storage market. I have seen my fair share of software-defined storage vendors out there. The likes of Nexenta, ioFabric, ProphetStor and many more have challenged into enterprise storage market, competing with the big irons, and have struggled to make an impact in their market footprint.
StorPool’s target market, seems to me, is the up-and-coming cloud service providers, the OpenStack ecosystems in both private and public cloud offerings, the web hosters. Many web hosters, SMB/SME cloud service providers that I know in Malaysia, Vietnam, Brunei, Indonesia, hate storage. As I mentioned at the beginning of this blog, storage technology is complex. Storage has been treated like a dark art, and once it is working, it is best to leave it alone.
StorPool managed services and pricing
That is why the StorPool managed services make a lot sense. “Leave the hard part with us, and you take care of your business” seems like their mantra to win market share. I love this approach and I hope they will grow more into the Asia Pacific market as well.
A quick check to their website reveals their pricing model (as shown below):
A night of margaritas
The first evening get-together after Day 1 was at Olla Cocina. I had a couple of margaritas (they were quite strong) and I was not much of a drinker. Coupled with jet lag, I was struggling to keep a straight mind, but I had a nice conversation with Boyan Krosnov, StorPool’s Chief of Product. I found out that he spent some time in Malaysia some years ago. It was nice to know that. All the best to him and Boyan Ivanov, StorPool’s CEO who were so passionate about their technology. It was wonderful to know them and hope to engage them soon in South Asia.