A Look at the Key Business Reasons for the Komprise Data Management Architecture & Why They Matter
We built Komprise based on customer feedback. As enterprises told us they were literally drowning in data, we realized there was a massive innovation gap. Most data management architectures were built more than a decade ago, back when data footprints were still relatively small and standard protocols for storage were still emerging.
Products to address different aspects of data management, such as Information Lifecycle Management (ILM), Hierarchical Storage Management (HSM), Data Virtualization, Data Analytics, and Storage Resource Management (SRM), were all too complex, too costly, too niche, too proprietary, and hard to adopt.
We designed the Komprise Data Management architecture to create an easy to manage, easy to scale, storage independent, no lock-in solution that delivers business results.
In this blog, I will walk through the four key business problems the Komprise architecture addresses and why they matter.
You can get an under the hood look at the pillars of the Komprise Intelligent Data Management architecture in a blog written by our CTO, Mike Peercy.
Four Key Business Priorities
The four business priorities that customers tell us they care about when it comes to data management are:
- Need to manage data at scale across current and future storage
- Need to manage data without impeding data access
- Need to migrate and move data without disrupting current users and apps
- Need to extract and harness data value, not just manage infrastructure
Let’s look at each of these in a bit more detail…
Business Need #1: Need to Manage Data at Scale across Current & Future Storage
(i.e. Simplify managing data without lock-in)
Data growth is exploding in nearly every industry. As technology becomes more sophisticated, the more data growth accelerates. IT budgets and resources are constrained and businesses want to improve how they store and manage data.
The good news: there are a lot more options than ever before to store and manage data – you have flash, disk, object, cloud, even tape as storage options at different price/performance points. You also have multiple backup and data protection solutions: from software-only to appliance-based and cloud-based approaches.
The bad news: this confusing array of options is often dizzying and overwhelming. How does IT decide which options to pick? How do they decide what data goes where? What if they change their minds tomorrow? How locked in are they to a given choice? How do they know how much each option is going to cost?
IT is left to figure all of this out on their own: somehow glean visibility into how data is growing and being used across their current silos of storage, somehow put data into the right place at the right time, and to report and assess whether their current strategy is working.
What is needed is a data management solution that works across storage without being locked-in. A solution that also enables you to scale up and down as needed without heavy infrastructure investments. Therefore, we created Komprise as a non-storage, software solution that works across all your storage with a scale-out data management architecture.
Watch this video of a customer who was able to leverage Komprise to analyze their data, create a go-forward strategy, and manage data across silos.
Business Need #2: Need to Manage Data without Impeding Access
(i.e. Don’t get in the way!)
One way to manage data across storage is to add another layer in front of the storage that you want to manage – this was the popular concept behind Data Virtualization. Companies have tried data virtualization via software (e.g. most recent example is PrimaryData), via the network (e.g. Acopia), via storage (e.g. cloud gateways, array tiering) or via metadata catalogs.
These approaches have never taken off… why? This is because customers do not want a “middleman” in front of all their data, especially hot data and metadata. Imagine buying expensive flash storage and then having to stick a data management layer in front of it that is brokering access. This layer now becomes a bottleneck, a chokepoint, and a source for lock-in.
How can you have a data management architecture that sits to the side of your storage, works with all your storage without creating lock-in, doesn’t sit in front of your important data or metadata, and still enables access to all the data? That is what Komprise does. We built the solution to be storage agnostic, work via open standards, and provide access without fronting hot data or metadata.
Read this case study of how an Engineering firm was able to transform its storage and backup architectures without engineers noticing any change using Komprise.
Business Need #3: Need to Migrate & Move Data without Disrupting Users & Apps
(i.e. Users shouldn’t have to lose access or look elsewhere or rewrite apps)
Archiving data has traditionally meant that the data either goes offline into a long-term archive such as tape or the data moves to another namespace such as object or cloud, which also involves a paradigm change from file to object.
Neither of these approaches are ideal because creating behavior change for users and asking them to look for data in multiple places is hard and it’s equally hard to rewrite applications from file to object.
This is why Komprise creates a redundant, transparent, hierarchical filesystem that preserves file access to moved data, even when the data is stored as objects in the cloud. These files show up in the original namespace, so if you moved data from a NetApp filer or from a Windows File Server, you will see all the moved files as if they were still on that filer. Users and applications maintain access to the files, exactly as they did before, without any disruption. You can also directly access these files from the cloud itself, or natively as objects, so there is no lock-in.
Learn how Boone County was able to handle the onslaught of bodycam and dashcam data by migrating cold data to Azure through Komprise — without any user disruption.
Business Need #4: Need to Harness Data Value by Finding Relevant Data
(i.e. Enable new uses such as big data analytics on data no matter where it lives)
Ultimately, businesses are storing data not just for compliance purposes, but also to harness the value of all the data. Yet, businesses often struggle to search and find relevant data easily — since data is often trapped in multiple storage silos and was created by multiple users, many of whom may not even be in the company any longer.
A key need for businesses is to have a simple way to search and find relevant data sets across storage pools. For example, imagine that you are an insurance company and you wanted to find all files related to earthquakes across all your storage. Imagine if you could easily search and create a virtual data volume of any dataset, even if it is strewn across multiple data stores, and then copy this data into Hadoop or some analytics engine for big data analytics. Komprise provides deep analytics capabilities through a distributed search framework on all metadata across storage. We are getting a lot of customer interest and usage of this feature in beta.
As Komprise continues to evolve, we are committed to continue delivering on these four business benefits.