As a Cohesity partner, I won the Best Cohesity Sales Engineer award in 2023. My team discussed, essentially, improving Cohesity reporting because that's a feature gap I feel could really enhance the product.

Reporting is quite limited and not customizable. Many customers can't access specific numbers or insights regarding their daily data protection, something competitors like Veeam integrate quite seamlessly.

Cohesity DataProtect takes hours to restore a single file, even a small one.

Currently, Cohesity DataProtect occupies the total size of the virtual machine. In my opinion, the solution should create a snapshot of the used space in the virtual machine, which would help the environment consume less space.

I would like to see the CDP features integrated into this solution.

Replication is working great, but with backup replication, you should have the option to link and RPO to the backup schedule. This would reduce the RPO in case of a disaster. It would be great to have something like the Zerto software.

The GUI is in need of improvement.

The upgrade path is not so clear because there are too many branches.

The tool could benefit from further improvements in AI features, particularly in areas such as a generative AI chat interface. This feature could help answer queries about the data being protected. One feature that would greatly improve our experience with the product is the ability to identify data redundancies through a central console. This feature would help us reduce storage costs by managing redundant data. 

I would like alerts if any local accounts on the platform are changed for security. If someone were to change an admin account password, I would like an alert raised for that, which is not available at the moment.

The one area where we would like to see additional improvements is with reporting. There are very granular reports from each grid. There are also some built-in reports within Helios (SAS Management Console) that offers basic reporting. However, for complex, large-enterprise environments, a more robust reporting engine that supports detailed audit reporting and enhanced ad hoc reports would be welcome. There are REST APIs which can be leveraged and most of the data we need is tracked and stored within the Cohesity ecosystem.

When a new release is launched, there may be some bugs that need to be addressed by the developers. However, they promptly resolve these issues, and the software runs smoothly once again.

We have two big clusters for 16 nodes for each one, and we are trying to develop the IPAM technology for a new cluster on a vault site where we can store a third copy of our data.

The software can improve upon reports because, nowadays, reports are easy to understand. We have a set of reports on Cohesity, but we wish to use them on-premises. We want a feature with a set of reports embedded with the software.

I would rate the solution an eight out of ten.

The platform has certain limitations, mostly stemming from the restrictions imposed by the VMware API and some other factors. One area where there is room for improvement is the tracking of replicated copies, especially in the context of remote archive duplication.

Although there are certain interfaces and reports within the solution that provide detailed information about features like duplication to archive, they are not well-documented. In my experience, I have used some of these interfaces and reports in the past to gain additional insight. However, it would be helpful if these interfaces were properly documented and made readily available, instead of relying on support calls and breaking down URLs to access them. This would make it easier for users to leverage the full functionality of the solution.

You need to put the full name instead of a few letters while searching for backup. It should also improve customization. 

I can't identify a specific area that needs improvement. 

They could always keep up with continuous interface refinements.

In terms of what could be improved, it just needs some maturity. But, it gets better with every release.

In terms of the solution overall, it's so far so good. I'll have to check with my team, however, so far, from my perspective, it's working great. We aren't missing any features.

In terms of what could be improved, their UI requires a manual refresh. It doesn't seem like it updates itself sometimes when you're moving from screens or waiting for a change to take effect, so you have to manually refresh. I would say that is an area for improvement, and I would also say that they have room for improvement for being able to roll back to previous versions after you've upgraded. They don't have a good rollback methodology to go back to previous versions if there is a problem with the new version.

Cohesity DataProtect can improve in the area of databases. We have had some challenges with both Cohesity and Rubrik being able to effectively back up in a timely manner the SQL Oracle workloads.

In an upcoming release, they should add more migration capabilities across public clouds.

DataProtect is a full-featured backup suite. Better support for legacy platforms would be nice, but we should be moving off those anyway. 

The marketplace app stability is still an issue at times. The company is very quick and hotfixes are released regularly. Their support is really fantastic, and I’ve only had one notable outage relating to an update. Ultimately it boiled down to a configuration error on my part, so I suppose I can’t blame them too much for that. 

Now that we have added additional nodes stability has been much better. 

The canned reporting could be improved. Many of the reports are basic, at most, and at times we had to open a support ticket to help find or develop a report for our needs. This is something that even our last solution had in place already.

Error reporting could also be improved. Rather than 'errorcode -2', it needs to be more descriptive or at least, provide KB articles related to the error or problem. This would save time in creating a ticket and try to describe to the support engineer what happened to cause an 'errorcode -2' that most of the time, we don't know either.

The product is still young and is continually improving. There have been a couple of buggy releases in the software that have caused us minor issues, so possibly some better QA before releasing would prove to be useful. We have also been inquiring about new app and functionality via the Marketplace that was teased months back but has yet to come to fruition. The apps are coming, but slower than expected. While we know they are working at getting more of these features out and doing so as fast as they can, a better product roadmap would be helpful.

Cohesity is working on this, however, I would like to see a more expanded KB section. They have created a few videos also, more would be great. Their documentation is also currently being expanded.

Cohesity needs more hardening, in terms of stability. From our experience, we noticed gaps in the product function and the solution guides and documents. Better documentation is needed because the lack of information has necessitated having our team depend more heavily on Cohesity support. We need to have a more self-sufficient system.

More automation is needed for easier management.

There have been a number of improvements since we purchased the product two years ago that has since made it even more attractive to users. Things liked education and training have been a greater focus. Cohesity now offers "Cohesity Academy" for education, and recently launched Cohesity User Groups for people to join other Cohesity users in discussing best practices, use case scenarios, etc.  

In terms of the product itself, there have been a handful of patches that have had to be applied outside of the GUI via the command line. It's very straightforward and easy and in the same redundant fashion, but not quite as simple as one or two clicks. I'm hoping that will decrease going forward.  

Technical support being outsourced is a downside but not different from every one of Cohesity's competitors. I was an early adopter when technical support was with actual developers and engineers, and not technical support staff. So, you can understand my point of view.   

Features I would like to see include a more visible data lifecycle monitoring tab, and an easier to read backup status dashboard. For example, I would lie to be able to see the status of all VMs for all jobs without drilling into each one.

Chat customer/technical support would also be great.

Quality testing of upgrades should be somewhat improved. There really are many upgrades for new features and I really like that. Nevertheless nearly every upgrade brings its own bugs. Cohesity support, however, is really good and they always know how to solve these problems. 

While support does ask the customer if they are allowed to restart service XXX or set parameter X to Y, they never explain if this affects your business or if there can be any further problems by settings the parameters. As a customer, you have to agree to something you cannot really estimate. 

The error handling could be improved. The errors shown on the GUI tend to be just what the log spits out and that can be quite confusing. The nice feature is that the logs are in the GUI, so you don't have to go hunting for them, you simply need click the drop down to show the information. There are a few cryptic error messages, thankfully the support staff is fantastic and knows what the errors mean.

There are a few user interface concerns with some of the functionality, such as what screens you get dropped back on when moving in between different areas. If you use the back button on your browser, you may not go back to the last page due to the coding. However, overall navigation isn't tough or complicated. 

They are really ahead of the curve concerning development and telling it to the market. We are a partner, and we know that sometimes a certain technology about which they have been telling a broad audience is not really production-ready, or it is in a feature release. There is always a difference between general availability and general deployment. Sometimes, we sell the feature, but it is not in the GA. It is in one of the future versions. That's one thing.

The other thing is that they lack visibility. When talking about backup, the first thing that everybody says is Veeam. If you are talking about the appliance part of the backup market, Rubrik is the first one they're mentioning, but a lot of the Rubrik customers don't know Cohesity.

I would like to see the management of multiple nodes inside a tenant. If you want to manage a Cohesity environment, you either go on a per-node base, and you can hop from one node to the other, or you need to implement their SaaS capability but it always needs an internet connection. We would really like to have some kind of local management entity so that we can see all the Cohesity nodes in a customer environment and manage those nodes centrally. That's something that we would really like, and they are really working on it.

Reporting is one area that isn't as robust as it could be. The next release supposedly improves upon that but I have not seen it yet.

We had real issues with restores on our older Red Hat 5.x systems, but we have upgraded those and no longer need support for that OS. 

Being able to take a backup of OneDrive and Sharepoint in Office 365 would be great. It is in the Roadmap, but we would like to see it in the next release of this solution.

This solution is only compatible with a limited number of storage arrays, including Pure and Cisco Hyperflex.

The cloud management solution needs a little more development.  Now it is really like the same dashboard than the one you are using locally.  They put some AI in the prediction  of when you will be out of space but I would appreciate more AI to discover malware (but is in the road map for version 6.4 I think)  .

Volume Shadow Copies can get very large on SQL servers so be aware of this if OS says a drive has a lot of space but everything else is saying that the drive is out of space. This mostly occurs when the backup job fails, which is very uncommon. We've run into issues where network segmented computers couldn't reach back to Cohesity over 445 and that has caused restores to fail. Split large Database/NAS/Phycial backups into their own backup job so smaller devices can stay on their backup schedules and backup more often. SQL and Physical (Block Level) backups need to have a Cohesity Agent installed (requiring a restart) and they have to be manually added into Cohesity but the process is still pretty simple. Cohesity cluster upgrades have not been as easy as a single button click but a manual process that requires upgrading each node in the cluster individually. The upgrades do go pretty smoothly though.

We have been facing challenges, particularly with HANA databases. Despite initial promises of delivering an agent within a few months, it has been over two and a half years, and there is still no indication of when the Cohesity agent for HANA systems will be available. The absence of an agent for replicated recovery makes it impractical for our current needs. We are highly dissatisfied with the level of support and functionality provided for HANA within the Cohesity product.

We would like it to have more capabilities for different kinds of PaaS or SaaS services in the Azure cloud, such as Azure SQL or Azure Cosmos DB.

Support needs improvement, but the product is fine when well deployed.

I think Rubrik and Cohesity are pretty similar in terms of the user interface and user experience, but I think Cohesity could add more platform support. For example, they could improve integration with DB2 or make the integration S3 object storage a little tighter. Some of that stuff could use some work, but overall, it's a relatively comprehensive platform.

There were a few things that required quick support calls in order to get to function correctly. We were told that the issues that impacted us were going to be corrected on the next release and would not impact other customers.

We would love to see direct compatibility with HPE Simplivity, which we also leverage.

Cohesity could do a little better job communicating features in new releases and helping customers to understand when features will be released. For example, I found literature indicating that version 6.3.1 would have support for Active Directory backups when installed. That feature appears to have been pushed to version 6.4, but is still available in 6.3.1 if you call support and have them toggle it on.

For Oracle DBs, we have observed a few bugs. This can be done better.

We'd like to see easy Administration and non-disruptive upgrades and node additions.

For a few engineering issues, we've observed delays and they are asking for an upgrade of the cluster. It would be good if they modified it with gflags.

Knowledge sharing has become easy by technical sharing and has been evolving drastically.

I would like to see minor features added to the user interface, such as the software version in the dashboard.

I would like to see Microsoft Exchange DAG supported as soon as possible. I have been told that it's on the roadmap.

• I thought physical RDM's were supported, but apparently that is not the case. 
• The rack mounting instructions did not appear to be correct, but we were able to overcome it.

DataProtect's recovery time is too high. It takes a long time to recover things from the cloud. I'd also like them to increase the capacity of the file server. 

We have had our share of minor bugs over the past two years and a few bigger ones. But I think it's because the product has so many features, and they try to add new stuff reasonably quickly. For example, Cohesity adds certain functionality in what they call a "feature release." This doesn't mean a long-term release. They added new functionality very quickly, and it was things we needed that were not yet available in a long-term release. When I say "long-term release," I mean one that is more static and gets updated once every few months. We wanted the feature, but we had more issues than some of us might've liked.

On the other hand, they managed to fix all the bugs without significant issues like data loss or anything like that. So it was more an inconvenience than it was a real problem. But if I have to have any criticism of them, maybe it's that. There are some minor improvements on SQL Server backups that we are waiting on, five services. We had one problem that will be solved in one of the subsequent releases, but those are small things. We are not waiting on any major stuff. The product currently meets all our backup needs.

We are waiting on the AIX agent to be released. We are also waiting for SQL Clustering to go from testing to live.

Add the ability to delete View Boxes and External Targets in the GUI. Currently, once you setup a View Box or External Target, there is no way to remove them (if you set these up for testing, for instance).

• Cannot exclude certain drives from VMs.
• Cannot back up just directories.
• You have to reboot physical servers after installing the backup agent BUT this is because of the CBT driver that it installs. I totally understand the need, but I figure it is worth pointing out.
• Only archive options are to the cloud or tape; would like to have a disk option.

The support team sometimes takes a long time to resolve issues, although, in our experience, these have been specific things like SQL blobs.

It can be that some Bugs will be fixed in future updates and you have to wait. But normally they will release a Hotfix for you.

One thing that would be great to see in the next release would be an complete restore for an SQL Instance (or multiple Databases). At the moment you have to restore one by one.

I can't wait for native bare metal recovery. It would also be great to add SQL Object Level restore, so user objects, tables, etc. can be restored without having to restore the entire DB.

It should have quicker restore times.

The solution's dashboard or UI can be enhanced and made more user-friendly than its competitors.

We've certainly had our share of problems with it. Since March, I think we've opened about 150 tickets with Cohesity for support for various issues; we're averaging 4 tickets a week with them. We're working through the issues that we have, but it seems like, on many levels, we're rewriting their code to make it work in our environment.

We, as a city, a local government — we have a lot of old legacy hardware, old operating systems, and old applications.

Commvault was pretty good about supporting those legacy operating systems that we still have in our environment, which includes things like Windows 2000 and Windows 2003. I understand that Cohesity has to keep up with the times and everything, but our inability to back up some of those older platforms is a pain point for us.

I'd like to be able to pause a backup while it's running and have it resume when I tell it to resume. Unfortunately, I can't do that today. I have to basically kill or cancel a job — that's one of my biggest pain points. I'd like the ability to backup clustered file servers, which I can't do today. I have to back them up as NAS devices basically, as opposed to Windows Servers. Those are my two biggest pain points. 

There is limited support for legacy systems, which is something that could be improved.

Navigation is sometimes a challenge from the aspect of drilling down into details which isn’t normally needed.

We don't love the Exchange integration, it works fine but requires a third-party application called Kroll. We'd rather have a built-in solution to manage Exchange messages/DB level restores such as what Veeam provides.

The pricing of the solution could be better, as it is quite high when compared with the competition, Veeam, Veritas and Rubrik. 

The solution offers many features, so my one reservation involves the pricing. 

The solution has a more complex setup than Veritas and Veeam. 

Currently, the product is lacking some features like BMR and backup for the cluster, but I think these are things that would be added in a few months.

I would like to see Cohesity have the ability to do a file level restore from a Windows server, which is storing data on dynamic disk.

Multi-Tenant for single cluster

We're still waiting for the ability to backup SharePoint Online & OneDrive but have been told it is coming soon. As we move towards a "cloud-first" implementation strategy for new solutions, these types of backup/recovery features will become more important.

We have Oracle databases running on Linux servers in our environment and we have found the Oracle backups were quite complex to setup. Simplifying the backup and recovery of Oracle would go a long way.

NAS incremental backup performance is a bit slow. This has apparently been fixed in version 6.4.0, but we waiting for it to go LTS before upgrading.

Support is good, but some functions are added pretty late, like vCloud Director support. This could speed up adoption for service providers.

Application integration.

• I would say they could work on the update process of the nodes. 
• They could also make it easier to try out beta versions of their software without the need of opening tickets.

• Bare metal backups of MS AD
• Microsoft Exchange online mailbox backups would be fantastic.
• Restoring OUs/individual users from the AD would be awesome.

A little more documentation on standard faults would help, as new features are being released quite often. 

Faster backup and restores, reduced costs of backups.

Since it is an evolving product, there are lot of frequent releases which require upgrades to be done every other weekend. 

At times, you may click the wrong server or click too fast and accidentally restore a cluster of servers. It would be very helpful to have a cancel option under restores. I have spoken with the engineers at Cohesity and this should be rolled out with the next release.

Better cluster support.

The one drawback of this quick innovation is that their units were shipped with outdated firmware, but this was quickly remedied, and they have made operational changes to reduce the likelihood of sending outdated firmware again.

Replication between two clusters should be included in the next few releases. There are many features which will be added in the next few releases.

Replication between two Cohesity clusters should be be improved in the next releases. Some workflows in the GUI should also be improved within next releases. Support for more vendors (databases) should also be delivered within the next releases.

Mature at a good pace.