Once you establish a replication job, when the file system is synchronizing its data to the target file system, it tends to create a balloon effect where it creates a file system that grows as the new data is synchronizing. It would be very helpful if the replication job could incorporate a self-cleaning job after the synchronization job completes to reduce the size of the file system in the target automatically. Very often, we have to run a script called “SavVolReclaim” to clean up space consumed while the replication is in progress. If the replication job hits its maximum capacity of 16TB while synchronizing, it stops the replication job. By adding this reclaim process natively, it would eliminate additional unnecessary work for the storage administrators.

It would be very helpful to get an automated report that shows you the size of the checkpoints and get warnings when a checkpoint is reaching either maximum capacity per a file system or hitting the ceiling on the SavVol pool consumption.

View full review »

There is no easy way to defrag a RAID group. It would be nice to be able to reduce the size of a storage pool if the storage is not needed anymore.

When you need to defrag a raid group, in earlier array versions of EMC arrays like the CX line all you needed to do was to right click the raid group and select the Defrag option and the array would defrag the raid group. Now with the VNX series arrays, if you need to defrag a raid group it is a multi-step process that is all manual.

1. You need to list all the LUNs that need to be moved.

2. Then create temporary LUNs of the same size and SP in a location not in the raid group.

3. Migrate these LUNs to another location outside of the raid group.

4. Then create the LUNs again back in the original raid group, this will move all of the unused space to the end of the raid group.

5. Migrate the temporary LUNs back to these LUNs in the original raid group.

With that being said, with the VNX array, you never have to defrag the space in a Storage Pool.

View full review »

EMC could have made the effort to spec the device appropriately. Instead, they low-balled the config in 2011 to make the price tag look alluring. Then, less than 12 months in, we needed to spend 35% of the cost of the entire array to put enough disks in the array to make it both perform and contain enough capacity, which only lasted until mid-2013. The device had multiple issues from 2013 – 2016 and numerous support calls to no avail, until I told them they would lose our business if they couldn’t identify the issue.

They finally spilled the beans, years after all of the issues, that the device was not spec’d appropriately for our environment. If they would have been honest about it, instead of cheaping out the config to make the solution appear appealing, this could have been avoided.

The competitor’s product was more expensive but would have destroyed the VNX’s capabilities. All they cared about was the sale.

View full review »

EMC VNX needs to support bigger SSD and the Next Generation EMC Unity does this.

View full review »

The management software used for the VNX is Unicenter. While it is an improvement over Navicenter, used in older EMC SANs, it still feels outdated in comparison to other SAN management software. Reporting is clunky and requires data collection to files, which then have to be retrieved and uploaded to a third-party site for processing before finally downloading the report.

View full review »

Storage is undergoing a sea change in affordable cost per bit, solid-state drives and new latency requirements for applications. Over time, VNX has become pricier than its competitors, and we have turned enthusiastically to Unity. While the VNX-F was released several years ago as an AFA in VNX trim, we still went with EMC XtremIO instead because of performance and pricing.

View full review »

Make major changes to the file services, Celera, do something like a lite version. It used to be more solid product before unifying Clarion and Celera.

View full review »

The CLI could be better documented, like with VMAX.

View full review »

I do not know, because all products function well.

View full review »

So far we haven't had many issues with the VNX. However, the Data Mover (that manages CIFS/NFS) part needs some improvements. EMC's new storage (Unity) has no DM's as the NAS functionality is all virtual.

View full review »

There is an easy replication process between distinct data centers via VPLEX and/or RPA. But I hope in the future that EMC/Dell could replicate this with other types of storage.

View full review »

EMC VNX needs to support bigger SSDs. Next generation EMC Unity will do this.

View full review »

The performance impact snapshotting had on the cluster was enormous. The entire cluster performance took a performance impact of about 30% as soon as we created one snapshot of all the volumes created on the cluster.

View full review »

It is not the easiest product to implement and manage, especially on the NAS side, but that is now fixed in its successor product from Dell EMC called Unity.

View full review »

I preferred HP EVA over Clariion for block due to the simpler GUI, but the VNX (that uses Clariion-like block) appears more complex due to the flexibility it ultimately offers. It has more functionality/flexibility which means a more complex interface. But, all things considered, Unisphere is a very good interface. It doesn't seem too fast though, but that could be the environment rather than the VNX product, to be fair.

View full review »

It only supports 10Gb copper, and cannot do 10Gb fiber. The system health screen could be more useful and show more important information. No support for flash drives.

View full review »

• Storage processor architect for real active/active HA
• ASIC based functions are needed

View full review »

I think one area where we would like to see improvement is better deduplication and compression. Some other vendors offer better deduplication and compression ratios and performance.

View full review »

I would like to see a larger SP cache and finer grained auto-tiering.

Based on our workloads, we see repeatedly in performance reports that the built-in controller (SP) cache of our VNX model is not sufficiently large, resulting in forced cache flushing.

The report states that this "can be addressed with a shared global cache in next generation VNX systems". But for this VNX generation (1st), the cache size results in an overloaded system.

Regarding auto-tiering, this generation VNX FAST algorithm moves data in 1GB chunks between tiers. With our predominantly transaction oriented workload (random reads) this didn't afford benefits.

So when we purchased new VNXs (same generation), we configured the SSD tier as FAST Cache, a large 2nd tier consisting of 10K disks with a small 3rd tier with NL-SAS disks. Had the block size been smaller, auto-tiering may have been more effective for our workloads.

View full review »

• Upgrade the GUI
• Full integration with VMware
• The Unisphere for VNX needs improvement as it is old and slow.

View full review »

The processing speeds of the Xeon controllers and scalability options are areas with room for improvement:

• Intel Xeon processors with under 2 GHz processing speeds could be replaced with more recent ones.
• Also, the scalability (connectivity between racks; daisy-chaining made more robust - basically, loops of 8 DAEs to be enhanced).

View full review »

VNX can improve by offering flexible upgrade options. It's not possible to add a single HDD to a current array and there are fixed rules to make upgrades.

When I work with other vendors like IBM, I can add a single disk to an array to get more capacity, it's simple. For VNX, you have to add a new array, not just a single disk.

View full review »

The device is nicely planned and user-friendly, but the administrative console (Navisphere/Unisphere) needs some improvement, especially on their Java-based GUI. The updated version of Java is not compatible. Also, the interface is delayed on updating its status (of snapshot).

View full review »

I can’t think of anything because they’re evolving more and more. They’re making headway into datacenter where non-structured data-type infrastructure around Hadoop, security, encryption.

View full review »

We get good performance from all our applications like Oracle DB, SAP, Sybase DB, Exchange, and SharePoint which are running on the backend of my storage device.

View full review »

SDS (software-defined storage) is so required in next releases because we are going to the next generation of IT infrastructure environment.

View full review »

• Caching
• Auto-tiering

View full review »

Poor connection to FC.

View full review »

As an entry level array, this is just right, but adding support for FAST-VP would be beneficial although that feature is available on the higher end VNXe3200.

View full review »

Due to very complex structure and sophisticated architecture, it needs experts for trivial troubleshooting also. It’s bulky and expensive. View full review »

They may change the disks from NL-SAS to Flash, but it would be costlier for customers.

View full review »

Main business applications run on it.

View full review »

Ease of use of the management software

View full review »

The next release is the Unity storage. That is the new family. Many of the features that are included in the Unity have improved over the VNX features. The VNX is a little bit old. What would help would be if the VNX had embedded encryption. That would be great.

View full review »

Source deduplication and compression will improve the TCO of these storage systems.

View full review »