IBM FlashSystem Room for Improvement

Abdul-Salam - PeerSpot reviewer
Sr. Manager - System Analyst (Datacenter Infrastructure) at Sohar International

Our model does not support compression or deduplication so it has less memory and the controllers are less powerful. We cannot get maximum capacity out of the storage. 

Bigger models support these capabilities but it would be nice if entry-level storage did too.

View full review »
Mike Bellemans - PeerSpot reviewer
Project manager at Sopraco NV

IBM could do more marketing and improve brand awareness. I had never heard of this product until a colleague told me about it. FlashSystem could add a few features, but it would probably increase the price. For example, Pure Storage offers instant snapshotting and partitioning. That would be nice to have, but I think the cost would go up. 

View full review »
Storemgr67 - PeerSpot reviewer
Storage Manager at a financial services firm with 10,001+ employees

In IBM FlashSystem, data reduction is an area with shortcomings where improvements can be made in the future.

In the future releases of IBM FlashSystem, I want to see synchronization in three-site replication specifically.

The response time taken by the technical support team of IBM is an area where improvements are required. It takes a lot of time for me to get to IBM's second line of support to get an answer.

View full review »
Buyer's Guide
IBM FlashSystem
March 2024
Learn what your peers think about IBM FlashSystem. Get advice and tips from experienced pros sharing their opinions. Updated: March 2024.
765,386 professionals have used our research since 2012.
Atif Najam - PeerSpot reviewer
Chief Information Officer and Program Lead at Gatron Industries Ltd

IBM FlashSystems is lagging in optimizing storage technologies, which can be improved. For example, deduplication with Dell EMC and their storages are far ahead of the curve regarding data storage deduplication. Deduplication is when you remove all the data duplicates from your system and save more data in a smaller physical space. If there are any duplications in the entire storage, they should be removed, and data can be saved. There are two types of deduplication, inline and post-save deduplication. Inline refers to the data travelling from the system to the storage to be stored. The system removes the duplicated part and then stores it to save space. Another kind is that you store the data first and then remove the duplications, so you will need a larger area to read and write. The inline deduplication needs to have global deduplication enabled.

View full review »
CM
Web Operations Manager at a manufacturing company with 10,001+ employees
  • The overall code.
  • The ability to have built-in reporting to understand the health of the array.
  • Stability within the code.
  • Clear upgrade paths for the code.
  • Support is horrible. It takes weeks to get simple answers. We had a LUN down and support refused to get on a WebEx without the array being completely down.
View full review »
Raanan Sitton - PeerSpot reviewer
Vice President Sales at BELOCAL LTD

The only issue my team faced was transferring the data from the old system to IBM FlashSystem, which is an area for improvement in the solution.

View full review »
AM
Network and System Administrator at TWD Technologies Ltd.

We need to spend more time with the solution in order to detect any shortcomings. So far, we haven't really seen any.

The design is a little old-fashioned and could be updated. The rack is very primitive and designed in an older style. 

View full review »
MM
Product Manager at ZELINKA d.o.o.

The solution's pricing is a bit high. 

View full review »
EMIL DOMINGUEZ - PeerSpot reviewer
HW Sales Representative at GBM (an IBM alliance Company)

Our customers have raised concerns about the limitations of the FlashSystem 5200 and 7300, which only offer a 32-gigabyte connection. This presents an issue for customers whose other infrastructure, such as old HBA cards, is limited to an eight-gigabyte connection. To maintain the recommended N minus one connection, the customer must use the 32-gigabyte connection, which is not ideal. To provide more flexibility, our customers would prefer to have a 16-gigabyte connection on these small boxes. The customer was initially unhappy with the situation because they had to replace all of the eight-gigabyte connections in their sixteen servers in order to migrate to a new storage system. Furthermore, to remain competitive, the customer had to be given the machines for free. While the customer accepted this, it was still a displeasing and uncomfortable experience for them. However, the machines were eventually sold and installed, leading to a satisfactory outcome. Despite this, the customer was still left with a feeling of dissatisfaction from the negotiation.

View full review »
PH
Service Delivery Manager at a computer software company with 5,001-10,000 employees

The basic setup can be challenging when it comes to certain IP addresses and the configuration of the IP. You have to go in to different menus to makes changes and ensure it is stable.

Ina future release, we would like to have a better way to install and to configure FlashSystem.

View full review »
JA
Head, Internal Audit at a financial services firm with 1,001-5,000 employees

We had issues when attempting to do a flash, we hope to resolve it soon.

View full review »
Storemgr67 - PeerSpot reviewer
Storage Manager at a financial services firm with 10,001+ employees

They can improve its initial configuration. The initial configuration is currently very difficult. There are multiple choices or alternative ways to configure based on the use case and what you are targeting out of the device, that is, more capacity or more performance. These multiple alternatives cause a lot of confusion.

They should increase the processing part of the nodes. Currently, you can cluster up to eight nodes. From my experience and the workload that I am facing in my environment currently, I would like to see either a bigger or stronger node or a larger number of nodes that can be clustered together. We formally communicated to them that we need to see either this or that, and they are working on something.

View full review »
AZ
Infrastructure Architect Supervisor; Solution Delivery Supervisor at a financial services firm with 1,001-5,000 employees

They can include Amazon file system S3 protocol in the upcoming releases. It is a cloud file system. IBM FlashSystem doesn't have this feature in the box for high-end or mid-range. We have got requests for this from customers because we need to use S3 for EDI application storage. 

At the beginning of every year, IBM releases firmware. When I find any bugs in the firmware during the year, I am unable to find any information from IBM regarding the bug. I need to open a ticket, and the IBM engineering team makes a patch only for me. This patch is not public. By creating a customized patch for a client, they don't really solve the issue for everyone. If multiple users have the same bug, IBM should upload the patch on the official website so that we can download it.

IBM FlashSystem has a monitoring tool in the box, but it is not advanced. I need a more advanced tool for more advanced equations and monitoring. All top three storage vendors, that is, EMC, IBM, and Pure Storage, don't have a powerful monitoring tool. To monitor our box to show the statistics for I/Os and latency, I need to pay for extra software. The built-in monitoring storage is not mature enough to handle all requests and generate all reports that I need.

They can include the functionality to stretch a cluster natively without using any additional boxes. In addition, there are some features that EMC has integrated with the box. These features are not available in IBM FlashSystem.

View full review »
AANKITGUPTAA - PeerSpot reviewer
Consultant at Pi DATACENTERS

In the next release having the next level of high-speed performance would be great. We should see an increase in the FC board to thirty-two G, sixty-four G UBS port for the performance to be readily available. 

View full review »
Janith Ranasinghe - PeerSpot reviewer
Senior Manager - IT at LankaClear

This solution could be improved by offering greater amounts of storage. 

View full review »
Storemgr67 - PeerSpot reviewer
Storage Manager at a financial services firm with 10,001+ employees

I would like to see an improvement in the handling of large amounts of rights. An automatic flash system that doesn't do compression or deduplication will flush through the rights directly from the host to the flash modules. It doesn't keep them in the cache. For compression and deduplication systems, they have to do compression, deduplication and the memory and cache for the controller. So they have to keep the data there otherwise you will find yourself stuck with performance issues.

View full review »
Pratyush Bhosale - PeerSpot reviewer
Chief Manager - information technology at UTKARSH SMALL FINANCE BANK LIMITED

It would be good to have historical graphs showing data for a month. Right now, only live graphs are available.

Space reclamation also needs to be improved.

If you want to expand, you cannot expand the disc enclosure. You have to buy a total individual node. Sometimes, this is difficult because we are just looking for capacity and not a node.

View full review »
HB
COO at a comms service provider with 5,001-10,000 employees

IBM gets a lot of input from the field, from system engineers, regarding what needs to be done and I believe they have a roadmap to try to constantly make the user interface more mature, as well as the options and tools. I trust they are doing good job on this.

I have looked at a few pages of a report I download and I saw a graph there regarding software-defined vendors. IBM is not in a good position on this graph. I know that they are working very hard on this, to make it much better and to get to a level where it's not only hardware but also software to provide a complete solution.

View full review »
IA
IT Manager at Def Industry

The solution's infrastructure technology level could be PCI Express 5 instead of PCI Express 4 for the next version.

In addition to SaaS-based enclosures, the solution's expansion options must include NVMe options.

View full review »
RV
UNIX Security Consultant at a retailer with 1,001-5,000 employees

The solution has a low number of NVME host attachments at 16 per IO group over the fiber channel. This is magnitudes lower than competing products. 

The 8.5 release for the 7300 and 9500 Flash Systems no longer allows IO group migrations. The replacement volume mobility is not as seamless as IO group migrations.

The Kubernetes CSI driver and the open-stack cinder driver still rely on SSH instead of native APIs for configuration changes. This reduces the limit of outstanding configuration changes that can be submitted to storage in bulk. 

The solution has not yet adopted Swordfish APIs and its SMI-S APIs are legacy and depreciated. Swordfish's are vendor-independent APIs made by the Storage Industry Association that allow you to manage storage no matter your vendor. These new generation APIs were released after ten years but IBM has not yet jumped on board. With a multi-vendor environment like ours, implementations are easier with universal APIs. 

Redhat Enterprise Linux clones such as CentOS, AlmaLinux, or Rocky Linux are not supported. All are binary compatible and should be supported because they are fundamentally the same product with different branding. 

It would be helpful to have a public page listing the minimum supported firmware levels for HBAs from different vendors. We have run into bugs with fiber channel cards that were solved with firmware updates. It was a laborious process to cross-reference vendor information so it would be helpful for IBM to provide recommended baselines for firmware. 

View full review »
Aamir Jameel - PeerSpot reviewer
Senior Systems Manager at Tapal tea

The interface could improve in IBM FlashSystem.

View full review »
SM
Head of IT Infrastructure at a insurance company with 501-1,000 employees

This product lacks some of the options we wanted. For example, expansion was difficult and it required a lot of patching to be done. It's not a seamless process because you need to do multiple modifications or alterations. There is a lot of effort required by the customer in order to expand the hardware.

Further expansion of our model is not feasible. As new models are released, expansion becomes difficult and you have to switch to a new model when you want to upgrade. This means that you have to migrate data between models. We want something where you can use for at least five years, where you can expand the hardware without the inconvenience of changing models. Ideally, the operating system and other applications would not be affected when expanding.

It would be helpful if the solution had built-in safeguards against security threats and malware, such as ransomware. Anything that can be utilized to enhance data integrity would be helpful. 

View full review »
Isanka Attanayake - PeerSpot reviewer
Manager - Information Technology Infrastructure and Development Support at Royal Ceramics

This solution is expensive compared to other systems. The interface of this solution could be improved.

View full review »
GM
BT Area Champion/Trainer at a financial services firm with 5,001-10,000 employees

Recently, we deployed SS9100. At the core level that is deployed on that storage, it is not stable. We had an incident not too long ago. Both controllers rebooted simultaneously, within 15 seconds. There was some threshold value defined in the core level, and the system exceeded that threshold value. 

We logged the case to IBM. IBM did internal checks, which we deployed. The permanent fix will be available in the first quarter of 2021. It seems to be an issue on IBM's side. Obviously, we were surprised by how both controllers rebooted. We faced downtime on our applications and on our services. 

The issue which we had recently faced relates to the core level. It should be first tested at IBM labs and then introduced for general release.

IBM should improve its data reduction development.

View full review »
Usman Rasool - PeerSpot reviewer
VP - Head Enterprise Technology Infrastructure at MCB Islamic Bank Ltd.

When you add more flash drives or more NVMe, that is the bottleneck or shortcoming with this type of storage. When you fill the first enclosure, you won't be able to add more NVMe until you add more SAS drives.

The security features can be improved such that the encryption does not affect performance in any way.

View full review »
MN
Infrastructure Solutions Architect at areebah

The pricing could be improved, but I think it's getting better and better with each version. 

IBM needs to implement NAS storage again, as this is a big flaw. Dell EMC is very good at this and if you compared them based on NAS storage, Dell EMC would win right away. IBM's solution for NAS storage is very complicated. We don't have a storage box that provides file sharing from itself, we have to put software on it and go through a whole complicated process. It should be simplified. 

View full review »
it_user674244 - PeerSpot reviewer
Design Engineer at a recruiting/HR firm with 10,001+ employees

There's always room for improvement. Adding deduplication would allow us to get more bang for our buck, basically.

View full review »
Branko Cirovic - PeerSpot reviewer
Storage Engineer at Comtrade Group

The solution should improve its pricing and the mechanism in the reduction pool. 

View full review »
it_user736929 - PeerSpot reviewer
Senior Systems Engineer & Support Contracts Manager at a tech services company with 51-200 employees

GUI interface should be enhanced more as there is some issues in copy services GUI interfaces should be enhanced as this will not let the customer doing wrong configurations

View full review »
MB
Chief Technology Officer at Inovo

ISCSI/ISER Bandwidth is at 25gbps. Would be nice if that can keep track with the bandwidth currently available in new dc networking deployments.

View full review »
RC
Hybrid IT Enterprise Executive at a computer software company with 11-50 employees

The marketing could be improved. Additionally, awareness in the marketplace could be improved.

View full review »
it_user672366 - PeerSpot reviewer
Senior Technical Specialist at a tech services company with 11-50 employees

I would like to see the snapshot management embedded within the GUI. Right now, the snapshot management, if you want to schedule, or do multiple scheduling in a day, you have to use scripting. Customers are not highly technical, and they really cannot do scripting. If IBM could include the snapshot management within the interface, then that would be good.

I know they have a flashcopy manager, but it is extra software, an additional license, and some customers don't like to add addition costs to their infrastructure. If IBM could create, or include snapshot management within the GUI, that would really be helpful.

I have a lot of customers who are really interested in the Storewize product and they are using snapshot management. It's really good to have this kind of investment, because this product is really good.

View full review »
it_user672375 - PeerSpot reviewer
CTO at CAS Severn

We think that IBM has to continue to invest in additional data reduction capabilities which are on their roadmap. Being able to use flash most efficiently, where the least amount of data is physically being stored on the V9000 is really where IBM needs to make additional investment. They are doing that.

View full review »
FS
Project Manager at Insirio S.p.A.

The cluster should be improved because non-disruptive failover was supported only on a few operating systems.

View full review »
it_user533121 - PeerSpot reviewer
Storage & Backup Transition Engineer at a tech services company with 10,001+ employees

The storage size vs the storage price needs to improve. The total storage capacity vs price is still quite high for the IBM Flash Array. It will sound better in the following years, as and when the price will be more accessible/economical.

View full review »
it_user211857 - PeerSpot reviewer
IT Program Manager at a media company with 1,001-5,000 employees

They could increase the maximum storage capacity so more disk shelves can be added.

View full review »
Nuno Chiziane - PeerSpot reviewer
Core Systems Services Consultant Infrastructure & Security at Asseco Central Europe, a.s.

The storage capacity of this solution could be improved. 

View full review »
it_user672384 - PeerSpot reviewer
Senior Technical Specialist with 51-200 employees

A lot of the beauty of the FlashSystem 900 is that it's very, very focused on what it does, and it does it well. I don't really think there are additional features required.

View full review »
it_user672363 - PeerSpot reviewer
Systems Engineer at a tech vendor with 11-50 employees

I would like to see the connection to cloud service providers, although I think that's already there. There's not much more that I can tell you. There are some issues on a firmware level. It's not very transparent. The option to connect to a Cloud storage provider seems to be available but there is not much detail on how it's implemented.

As a field engineer I’m used to how the systems work. Some customers however have expressed to me , well for lack of better word “anxiety” at not being able to see more of what goes on in the background (as an option). They would like a more detailed progress of the firmware upgrade process. (i.e. NVRAM,controller, etc…).

AHV is Acropolis Hypervisor – A relatively new Hypervisor, robust and stable as VMware vSphere, has built-in advanced analytics and powerful operations, Self Service Portal and components for DevOps included, managed by a single pane of glass (Prism) via HTML5 and it is free of charge – That is why Nutanix is so advanced and revolutionary.

The challenge here is in order to replace massively Install base of VMware vSphere they need to expand their ecosystem with ISVs (i.e. F5, PaloAlto, Citrix, TrendMicro etc.) to provide a wider range of solutions running as a virtual appliance (vApp) just like VMware did years ago.
Currently there is an ecosystem but I think it is around 30-40 providers.

View full review »
it_user211857 - PeerSpot reviewer
IT Program Manager at a media company with 1,001-5,000 employees

ACLs should be more compatible with other NAS-based platforms. Some of the platform limitations need to be addressed.

View full review »
SemyonOsadchiy - PeerSpot reviewer
Commercial Director at MUK

The setup could be improved.

View full review »
DP
General Manager at a tech services company with 11-50 employees

There should be more licenses included in the standard price.

Additional licenses might be added for the fundamental licenses, such as those for copying and flash copies.

View full review »
ST
Cloud Engineer at a tech services company with 51-200 employees

IBM's solution has come a long way and has had different milestones/features have been introduced. I would position the 7200 in the upper midrange class as it has lots of features - more than, for example, EMC Unity/VNXe. It doesn't lack anything, really. One could argue that NAS or S3 is not available to it, but I prefer other types of storage optimised for that job.

It is slightly more expensive, however, it all depends on your supplier. Licenses are volume-based. Larger companies with more TB usually are better off as the price per TB decreases the bigger you go. I would definitely recommend this platform!

View full review »
GA
IT Manager at a financial services firm with 201-500 employees

Deduplication feature is not mature and it is significantly increasing the latency. Additionally, the product has limitations in the software features and granularity. For example, data reduction feature cannot be aligned at the logical unit (LUN) level. It is aligned at the pool level. However, I expect to create some LUNs with deduplication and HW compression while some other LUNs with HW compression only. The other limitation is available at the remote copy feature. You can not create one-to-many scenarious for the LUNs which is created as the Hyperswap LUN.

View full review »
it_user672393 - PeerSpot reviewer
Systems Architect at a tech services company with 201-500 employees

I can't really think of any specific improvements which I would like to see.

View full review »
it_user523167 - PeerSpot reviewer
IT Specialist at Slovenská sporiteľňa

It has room for improvement in the area of stability. The firmware is upgraded frequently. We need to update it maybe three or four time a year because of the new features and bug fixes. On one side, it is good; but, on the other side, it is not good because we need to upgrade within one year or they will not support the firmware. It would be better if they supported what we already have for longer.

View full review »
it_user672354 - PeerSpot reviewer
Storage Administrator at a government with 1,001-5,000 employees

With regards to the IBM V7000 storage system, where we have multiple tiers of storage, a heat map would show I/O distribution across the tiers of storage. I'd like to see some type of heat map reporting on the management console. It would be nice to see something like that. It may already be in the new release, but I'm not sure. The current release may have that already.

View full review »
JS
Deputy director at a tech services company with 10,001+ employees

The GUI for monitoring performance metrics could provide better visibility. For example, it doesn't let me segregate the IOPS per volume.

View full review »
MJ
Technical Support Manager at a computer software company with 1,001-5,000 employees

The price is very costly, which could be improved. In the next release, I would like to see some integration with VMware so that storage can be managed directly from a single pane.

View full review »
RC
Hybrid IT Enterprise Executive at a computer software company with 11-50 employees

Cloud file sharing is an area that needs improvement.

View full review »
BB
Deputy Chief Technology Officer at a comms service provider with 51-200 employees

The stability of the solution isn't great. We have had a lot of issues with discs over the years.

There should be better integration with utilization platforms.

The pricing needs to be more competitive.

View full review »
PV
Senior System Administrator at a tech services company with 1,001-5,000 employees

It could be easier to implement.

This is an expensive product and if the price were reduced it would be better.

View full review »
AR
CEO at a government with 1-10 employees

Enterprise data storage needs improvement. They should create a feature for data and file storage. 

View full review »
RR
CTO at a tech vendor with 51-200 employees

The support is simply not there, so it needs to be improved.

This solution needs a management console where we are alerted to issues and can report them, or escalate them through email or another method. If something happens to our storage, for example, then we will be notified, and we can report it through the console. 

View full review »
it_user730569 - PeerSpot reviewer
Senior Systems Consultant at a tech services company

The replication of data between IBM Storwize arrays in V1 of the product had some major limitations. A lot of the limitations were overcome in V2 of the product. Some improvement can still be made in scalability using clustered V7000 nodes.

One area for improvement is in the GUI, where host clusters are not properly dealt with. With Hypervisor host clusters, all hosts must see the same volumes in the same order. Using the concept of a “host-group” has been around (even with IBM) for many years, so why not with the V7000?

View full review »
it_user732744 - PeerSpot reviewer
System Administrator at Dhaka Bank Limited

It should increase sociability with respect to number of disk shelves and capacity.

View full review »
it_user672345 - PeerSpot reviewer
Senior Technical Specialist at a tech services company with 11-50 employees

There is a need for a little bit more of a drill down in the GUI part. There is a lot that they actually did, which is already available in the GUI but still, there's a lot that can't be done. So, I would like to see more improvements to be brought about in the GUI.

View full review »
it_user264465 - PeerSpot reviewer
Arquiteto de Soluções at a tech services company with 51-200 employees

Because the product is already mature, it doesn't need any improvement.

View full review »
RP
Senior Technology Engineer at a consultancy with 10,001+ employees

Replication features need improvement. Currently, they are there in the product, but I'm not sure as to how it works exactly.

They are other IBM products that are better than this such as IBM XIV and IBM disk storage (the old one); they have better performance.

View full review »
it_user674226 - PeerSpot reviewer
Senior Solutions Architect at a tech services company with 11-50 employees

They've already got one. The clustering host came in 7/8 code, so that's coming in. I would like to see deduplication come very soon. That would be really good to have deduplication built in, not only with compression, but also have deduplication with it.

View full review »
LS
Systems integrator at a government with 5,001-10,000 employees

I would like to see bigger modules. That's about the only thing that needs improvement.

View full review »
it_user671418 - PeerSpot reviewer
Enterprise Architect at a tech services company with 11-50 employees

I'm not sure that I want to see any additional features. I like the way they have it. It's simple. It gives you the low latency that you need, so I'm not sure if there are any features that I would be looking for.

I would like to have a larger disk. Right now, you can get 57 terabytes in a shelf. Once they get the larger disk and you get larger capacities, it'll be even better.

To be more specific: The FlashSystem 900 is perfect for providing pure speed and low latency. I am happy that the product doesn’t have Storage efficiencies built into it natively. You can Virtualize the FS900 with the IBM Storwize family and gain storage efficiencies when needed. Today the largest FS900 MicroLatency module is 5.7TB or 57TB per shelf. It will be great when the larger MicroLatency modules are realesed to provide density.

View full review »
JL
Director at a financial services firm with 10,001+ employees

The product needs to improve their scalability and the price.

View full review »
JV
IT Director at UAGM

Sometimes the performance is effective but it gets resolved in the process.

View full review »
ME
Technology Supervisor at a media company with 1,001-5,000 employees

The solution is quite expensive. That's one of the downsides to using it. 

View full review »
it_user672396 - PeerSpot reviewer
Storage Administrator at Sprint

There is a need for more help with the commands because, from version to version, they change; so what I used to use back in the different version, I can't use now. Thus, it would be helpful if they improve that aspect.

View full review »
it_user671433 - PeerSpot reviewer
Programmer Analysts at a government with 1,001-5,000 employees

I had wish lists, but I found out that they have already been completed.

At our current code level, when I extend any volume that is using Metro Mirror, I have to remove the Metro Mirror relationship. Once the Metro Mirror relationship is removed, I can extend the volumes and I have to re-create the Metro Mirror relationship and sync the entire volume.

For example, if I have a 1TB volume and I want to extend to 1.5TB instead of 0.5TB being replicated using Metro Mirror, 1.5TB has to be replicated.

The next version will include an update for this problem.

View full review »
it_user671412 - PeerSpot reviewer
Network Engineer at a tech services company with 11-50 employees

As far as for what we use it for, they have got it spot on for our market. I'm sure others might have some additions, but our market is more in the SNB space. It is everything we've ever needed.

With the latest releases of software, it's pretty robust. They've pretty well nailed it. I don't see any further weaknesses with the interfaces or anything.

View full review »
it_user671409 - PeerSpot reviewer
Storage Engineer at a financial services firm with 10,001+ employees

I would like it to be more scalable.

View full review »
KE
Chief Technology Officer at areebah

The generic functionality of IBM FlashSystem, IBM always dismisses using file share or sharing protocols inside their storage hardware, and they only focus on the block-level storage.

View full review »
MN
Software Developer at a tech services company with 11-50 employees

The deduplication and compression ratio is not very good. It's not reaching a very high ratio.

I would like to see the replication or synchronization between the storage to be IP replication. Right now, the IP replication is only supported in this model. So, the pure IP will be okay.

View full review »
BS
NDT - TECHNICIAN at a paper AND forest products with 1,001-5,000 employees

We use some open-source tools for monitoring, such as Grafana and it should be bundled along with IBM FlashSystem.

View full review »
it_user672354 - PeerSpot reviewer
Storage Administrator at a government with 1,001-5,000 employees

Include an option to upload the support package to the IBM ECuRep when opening an IBM PMR.

View full review »
it_user738690 - PeerSpot reviewer
Gestor de Capacidad at a logistics company with 10,001+ employees

Of course the cost, it's expensive compared to other solutions.

View full review »
it_user583515 - PeerSpot reviewer
IT Director Of System & Security Dep at a transportation company with 1,001-5,000 employees

As I said When you Sync V7000 to Another V7000 that are located in my case more than 10km and with DWDM as dark fiber , first you need license for long distance from brocade.
Than you can change the buffer credit on the SAN switch to at least 49 . the point is that you need to test the parameter to get the best value .
This depend on the line you have and also check if there is any limitation from your DWDM Provider.

The problem is that if the buffer is not configured correctly this increase the latency because of this delay and the retransmit .

View full review »
ST
Senior System & Storage Engineer at a energy/utilities company with 1,001-5,000 employees
- CLI though intuitive, no other API available - Lacks scheduling or a cron - Has no built in short term performance graphs - IBM TPC overexceeds the montoring needs, had to fall back on Stor2RRd - Support response times are bad - View full review »
it_user674229 - PeerSpot reviewer
Storage Admin at a tech services company with 1,001-5,000 employees

There is a limit on the flash memory for fresh copy, i.e., a memory limit. So there are limits within the software, for buffering and then caching. We want to see those items increased.

There are a lot of bugs and they are still adding features in the product. The system has matured from 6.X to 7.X. It has improved a lot, but it still has room for more improvement.

View full review »
it_user526413 - PeerSpot reviewer
Senior UNIX Systems Administrator at a financial services firm with 1,001-5,000 employees

The several major crashes resulted in the entire weekends being lost in rebuilding the array and then for restoring the data. This was caused each time by a memory card failure and the array's inability to use a spare card. The array level RAID does not seem available.

View full review »
it_user739677 - PeerSpot reviewer
Systems Integration Expert & IT Consultant at a tech services company
  • The IP replication
  • Events/log analysis tools
View full review »
it_user531666 - PeerSpot reviewer
Senior Analyst at a tech services company with 10,001+ employees

The two products should be unified and have a service processor to restart devices remotely. For example, NetApp has devices which include a service processor in nodes, which could be used to restart the nodes remotely. IBM should implement something like this.

Both products are basically for block-level access. There should be a series that also has a file level and block-level access together, which would make it unified.

View full review »
MD
Storage Administrator at a tech services company with 10,001+ employees

A big area for improvement is that the data reduction pool feature is not recommended for use in a production environment because it has stability and performance issues. There are also some issues with updates when you have different versions installed, which makes it quite difficult for the administrator to choose which version to install.

View full review »
it_user541197 - PeerSpot reviewer
General Manager Information Technology Department at a university with 501-1,000 employees

Product support is restricted to IBM only. It must be decentralized to IBM partners as well.

View full review »
it_user672369 - PeerSpot reviewer
Solution Architect tct at a tech services company with 51-200 employees

The HSM management console GUI still need some enhancements. It is still a bit awkward in navigation, and some functions/features are not yet fully integrated into the GUI support, but only via CLI. We also need to get reporting down to individual pool and volume levels, rather than just overall system levels. Customers need a more granular view of their individual business usage as well.

View full review »
it_user672381 - PeerSpot reviewer
Head of IT infrastructure at a financial services firm with 1,001-5,000 employees

I don't know about improvements. Maybe having independent snapshots. For now, when we make snapshots one by one, they depend on each other. If we delete the first snapshot, then we can lose the others. We really need functional snapshots which are independent of each other.

View full review »
it_user671430 - PeerSpot reviewer
Enterprise Architect at Horizon

I don’t really have a feature that I’d be using. Other storage technologies such as 3D drives or 3D flash would be cool to see, but for the users who I’m selling to, I don’t have this to offer. I don’t think that there is any improvement required.

View full review »
KV
Director Technical at a tech services company with 11-50 employees

There is always room for improvement, but IBM is less interested in on-premise storage or on-premises solutions going forward. 

They're highly focused on the cloud. I don't see IBM being a very major player, the way that they used to be because they are moving away from this and are trying to move all their customers to the cloud.

Nothing really comes to mind for needing improvement. Some years ago, there would have been an answer to what could be better about this product, but nowadays, virtually all of the companies are meeting all of the features.

More and more, we would like to see how it's easier for the solutions to be bought by the customer more on a pay-per-use basis. That is certainly an improvement.

The customer's expectations are what they get on the cloud, they're expecting even in the on-premises deployments, going forward. 

They want to pay-per-use and not own and get stuck with what they're buying. 

They want flexibility.

IBM does that in a few products, but then more and more you see the business model changing towards that. We'd like to see that in all IBM products.

View full review »
AJ
Solution Architect at a tech services company with 501-1,000 employees

They should reduce the cost of the solution and offer smaller storage sizes to target small to medium-sized businesses.

The solution is not easy to implement. It takes a lot of time to study the product and it's a little complicated in general.

The solution should offer integration with the cloud on the next upgrade. It might already be there but we have not integrated that, or gone for that upgrade. It would mean we could keep a copy of our data in the secured cloud, whether it's Amazon or Azure. 

View full review »
it_user672390 - PeerSpot reviewer
San Admin at a healthcare company with 1,001-5,000 employees

There is need to put a SVC in there.

View full review »
EO
Systems Administrator at a tech services company with 11-50 employees

The ease of installation should be improved. We had issues with the configuration model. 

In the next release, there should be and flash and caching features. Customers also have problems accessing their files from the storage. That's what they usually complain about. This is something they should improve. 

The main issue is the speed in terms of accessing the data. That is the customer's big complaint. They also complain about the speed of the hard drive.

View full review »
it_user693222 - PeerSpot reviewer
IT Specialist / IBM Software and Hardware support at a tech services company with 201-500 employees

This storage is factory configured so it's very difficult and expensive to add more flash modules (disk capacity).

Scalability is the main area which needs to improve.

View full review »
it_user672357 - PeerSpot reviewer
Tactical Expert at a tech services company with 201-500 employees

I really don't have any preference for it, this is very easy to install and is actually a customer setup. I've talked to many customers, and all of them agree that "This is so simple, so easy, and straightforward", so IBM has done a very good job so far.

View full review »
SY
Technical Manager at a manufacturing company with 11-50 employees

The solution is not easy to use and could improve.

View full review »
AH
System engineer at a tech services company with 11-50 employees

The Data Reduction Pools (DRP) support could be better.

View full review »
it_user732741 - PeerSpot reviewer
Sistem Yönetimi Uzman? at a healthcare company with 10,001+ employees

Deduplication and DRC synchronization. Deduplication is small, limited. We need better deduplication and disaster recovery.

View full review »
DP
System Administrator at a tech services company with 11-50 employees

The data reduction pool feature sucks and is not recommended for use with heavy workloads. I also understand that the APIs don't work well and need improvement. In the next release, I would like to see a better monitoring system for metrics.

View full review »
IE
Senior Systems Engineer at a tech services company with 1,001-5,000 employees

Their technical support needs improvement in terms of reachability for the clients and response times. They should be more responsive and have more online platforms for support. 

They should make more technical information available online. There could be some kind of documentation community.

View full review »
CC
Technical Head at a tech services company with 51-200 employees

Flash Manager needs to be included.

View full review »
CC
Technical Head at a tech services company with 51-200 employees

When you provision a datastore auto-format takes a long time .. to disable is you have to redo the whole process. this is a bit difficult if you want to make adjustments on the datastore

View full review »
Buyer's Guide
IBM FlashSystem
March 2024
Learn what your peers think about IBM FlashSystem. Get advice and tips from experienced pros sharing their opinions. Updated: March 2024.
765,386 professionals have used our research since 2012.