Our clients say
Given the exponential increases in human knowledge across a myriad of scientific disciplines, we found ourselves facing an unprecedented data explosion. Our lab uses a combination of recombinant DNA, DNA sequencing methods, and bioinformatics to analyze the structures and functions of genomes. As knowledge in field of genomics increases, our datasets and analysis result have increased exponentially. We found ourselves at a crossroad in terms of how to mitigate this data challenge. We sent out a request for proposals to many of the top vendors and they returned with some very interesting and cutting edge solutions that would stem the data flood, albeit with a six figures tag. However, I posed one simple question to the vendors, “How will this scale once we pass the inflexion point on our data growth?”. They returned with several upgrade paths that we could transition to over the years which would require us to migrate to newer software versions and hardware architectures. This was a bit of a sticking point for us. Software upgrades? Yes. Entirely new architectures? No. If many scientific research infrastructures operate like NASCAR, we are more akin to the Pikes Peak International Hill Climb or Formula 1 racing. Yes, we need performant storage solutions but our path is not fixed. Our data will take us around curves, up steep hills, and down long straight ways. We need a solution that can support our work no matter how our research directions change. We need fast storage for current experiments and reliable long term solutions for the petabytes of data that we will store, forever. This requires us to have an agile and dynamic infrastructure in place.
I had looked at MooseFS several years ago and I liked the replication methods and trash retention system. However, at the time, our existing solution was tied to the hardware manufacture. When I looked at MooseFS this time, their new features such as Atomic Snapshots, Redundancy with Erasure Coding, Fast Disk Recovery, and Fault Tolerance, made me have to get them on the phone! They explained their solutions in detail and let me set up clusters in our own data center to see for ourselves just how powerful their solution could be. I had grown tired of multi-level disk health checks (smartd, then vendor tool 1, 2, … n) along with lengthy raid rebuilds and fearing failure of yet another disk in the rebuilding raid array. Creating backup solutions for petabytes is a challenge. Yes, one can mirror, deduplicate, compress, or (try to) write to tape but, I wanted a simple and easy online solution which MooseFS provided. In active redundancy scenarios (1K+ virtual machines), their Atomic Snapshots enable us to do this while also keeping coherence with replications. They had me sold and we have been using their products ever since. I can add on any hardware that I wish. Data on failed drives is automatically shuttled to other disks so I can hot swap disks from my multiple 90+ disk arrays almost at will. The Redundancy with Erasure Coding allows me to minimize space considerably. Did I already mention that I can add any device at will? No vendor tie in. Use the disks and architectures that you want whether cutting edge and expensive or cheap and after-market. Whatever you need to use for your infrastructure, we have found MooseFS to easily step up to the challenge.
Our research areas include Statistical Genetics, Genetic Epidemiology, Longevity Genetics, Cardiovascular Disease Genetics and we partner with numerous studies, academics institutions, and medical professionals around the world. During this period of COVID-19 pandemic, we have been tasked to double our research efforts. We have to plan for new samples for antibody and their cytokines even as I write this. Our storage requirements have doubled in the span of a week. The MooseFS team reached out to us and asked if we were involved in COVID-19 research. When I confirmed this, they gave us additional tools and extended our license to enable us to face these data growth challenges, today. That is an experience that I never had with our other vendors.
I highly recommend MooseFS if you are looking for a new, dynamic, and cost-effective storage solution.
— Steven Harris
Director of Information Technology
High Performance Computing
Division of Statistical Genomics
Department of GeneticsOne of the TOP 10 Medical Universities in the United States
This US Medical University uses MooseFS since: 2019
At virtion we offer hosting and managed services to our customers and providing stable services to our customers is one of the most important things for us. We started using MooseFS in 2014 because we were looking for a reliable and easy to manage storage system for certain use cases – e.g. backup, archiving, file storage and much more.
After we evaluated a few Software Defined Storage / Distributed File Systems, we found MooseFS Pro. Our intensive testing showed that it was the most reliable one, and it still is – it requires very low maintenance efforts while delivering a very good performance and – much more important to us – stability.
The ease to manage a cluster in our hyper converged infrastructure – adding or replacing machines and disks over time, on the fly – led us to use MooseFS Pro for more and more use cases and customers supporting the ever-growing storage demand.
We have upgraded to MooseFS Pro 4.x recently, therefore we can start using a new feature of Erasure Coding available now which will allow us to save space on the cluster.
Michael Kutzner
CEO
virtion GmbHvirtion GmbH uses MooseFS since: 2014
Our company got familiar with MooseFS at the time when company’s infrastructure started to grow in size exponentially. As more and more virtual servers were introduced the more public cloud limitations we faced. One of the most frustrating disadvantages was SAN storage failures and worse than expected I/O performance. We made a decision that we would better off by using Infrastructure as a Service solution where we would have self-support and decision-making possibilities. Our choice was a Hyperconvergent server solution combining computation, network and storage into a defined hardware units – something like a big-scale hyper modules. This enabled infrastructure to become hot- swappable set of elements with minimal downtime and highest fault tolerance and also greatly simplified scalability of the whole system.
In order to implement such vision, we had to choose a proper storage software which played a major role in this scenario, however most of the current modern solutions have one or more critical disadvantages in elements we took into consideration – they either lack in area of performance, high- availability, scalability or the cost. As an IT engineer I was really surprised that MooseFS was the best choice in the market after trying out and comparing multiple well-known enterprise-level solutions as it was certainly the easiest to implement, most flexible (especially for having ability to have different type of drives on each hardware unit), most tolerant to faults (f.e. unexpected power-offs), and self- recovering meaning that maintenance was a trouble-free process. I am also pleased that each new hardware unit brought into infrastructure increases the overall performance this way showing a great potential in scalability. Last but not least, I really enjoyed having great support on MooseFS and having every one of my question answered in a swift and detailed manner even before buying the actual product – a thing I never expected to find after many frustrating years dealing with corporate “premium” support lines.
So far, our choice for MooseFS seems a right decision as we have achieved the goal we desired – to have an infrastructure of high performance, redundancy and easy maintenance while keeping cost-efficiency in mind. I can honestly recommend MooseFS, especially from the perspective of IT engineer as it over-delivered my expectations and is an example of software done right.
Hotelston Ltd. uses MooseFS since: 2018
We have about 1000 webinars per day and of course our clients want to see their events recorded. For us it means lots of media files should be stored for on-demand watching. We have tried many solution but got REAL stable results with MooseFS only.
We have tested MooseFS cluster two weeks. We have placed and then read almost 100 millions small media files… And MooseFS has passed all our tests well after tries to break it. We got no luck, MooseFS got respect! Thank you guys for great product!
Sincerely,
MyOwnConference teamAkovana LLC uses MooseFS since: 2018
We started deploying MooseFS Pro as a Distributed File System on all our storage servers. This eliminated the Single Point of Failure we previously had, as now copies of each data block resides on multiple servers. With this new storage model we not only have redundancy within each server via mirrored disks, but we also have redundancy at the server level, meaning that an entire server can fail without loss of data access. Furthermore, in case of a power outage, the recovery time is lowered to mere minutes instead of hours or days.
Furthermore, this model allows us to add storage capacity on the fly, set higher redundancy (ability to store more than two copies of the data) and also distribute our data geographically in order to mitigate against local problems or to meet new regulations – EU data could reside in a EU server and be served locally.
We are now ready to grow on much more solid foundations and although it was a challenging past few months, it was definitely worth the effort!
PolarisMail Inc. uses MooseFS since: 2018
For data storage, Dedikam uses MooseFS, a robust, stable, redundant and high-performance file system. MooseFS is distributed under two types of license, open source GPL and a PRO version. Dedikam uses the Pro version, which makes it possible to have a system in high availability.
MooseFS checks the integrity of the files by a hashing algorithm and constantly checks the different nodes of the cluster to correct any type of anomalies. All Dedikam data is at least double replicated. If a hard disk or server fails, the data is automatically replicated to other servers and hard disks. Thus, user data is always available even in case of hardware failure. MooseFS makes it easy to add additional disk space (adding a server or hard drive) without any downtime.
The MooseFS team is very professional and responsive. In case of a technical problem, there is always a competent person to assist us or help us to find a solution quickly. We have been using this solution for 10 years so we can say that the system is reliable, robust and secure.
Integrity and security of data is fully assured with the MooseFS file system. We are very satisfied.
Dedikam uses MooseFS since: 2008
Ipla is the Internet TV broadcaster streams on-demand content 24/7. It’s possible with the underlying MooseFS storage build with commodity hardware. We appreciated that the system is ready for unexpected peaks with its parallel, distributed design. The show must go on!
Ipla uses MooseFS since: 2008
Our core business has been relying on MooseFS since 2005. We keep a long history of internet traffic and currently have more than 6 petabytes of data. Our storage increases at 300,000 events per second. With MooseFS, we are sure our customers get their reports on time.
Gemius SA uses MooseFS since: 2005
As a researcher I need to deal with large sets of 3D points clouds. Hell of data. MooseFS helps me handle them, I can verify my ideas and algorithms efficiently. Atomic snapshots rocks! In addition I’m sure my work results are safe.
Mandala Robotics uses MooseFS since: 2017