Loyal readers, it's that time of year again when we look back to see where we've been and look forward to see where the data storage industry is going. Every year, I try to predict technology winners, losers and those stuck in the middle. For those of you that want to look at last year’s article: Review of 2012 and Predictions for 2013
First let’s see how I did last year.
Looking Back: Predictions from 2013
1. My first prediction is that sometime in 2013 we will find that PCIe-4 is going to be delayed--probably till 2016. This means that technologies such as 100 GbE will have to run in 16-lane PCIe slots. Without 32 lanes you cannot have a dual port card.https://o1.qnsr.com/log/p.gif?;n=203;c=204660761;s=10655;x=7936;f=201812281257540;u=j;z=TIMESTAMP;a=20400368;e=i
I got this one correct. The latest from the PCI-SIG web site says that PCIe 4.0 available in late 2015. We have had a one-year slip. In addition I was told at Supercomputer 13 that vendors are making 16 lane slots.
2. There will be continued market consolidation among SSD vendors. The market has too many vendors with limited technology differentiation and with high marketing and sales costs. This model does not work nor can it be sustained, given the development costs.
I also got this prediction correct, with WD buying two SSD companies, Cisco buying Whiptail, and poor OCZ looks like it will be purchased also. Combine this with the fact that few to none of the SSD-only companies are doing well financially, and it appears that we’ll see more of this consolidation in the future, especially in 2014.
3. 10 GbE per port switch prices will finally be priced reasonably. The number of 10 GbE chipsets is growing fast. Soon, the market will be flooded with new chips and this will drive the prices of switch ports down.
The cost of 10 GbE chipsets has dramatically dropped this year and both NICs and switch ports are very reasonably priced compared to a year ago. The same is starting to happen with 40 GbE.
4. In the enterprise, we will finally see a significant movement to 2.5 inch disks rather than 3.5 inch drives. There are lots of advantages of 2.5 inch drives over 3.5 inch drives in terms of watts per IOP and watts per MB/sec. This is not to say that I am predicting the death of 3.5 inch drives – just movement to reduce the market share for 3.5 inch drives.
The disk drive market has been slow to move, especially in this direction. There has been some movement, but nowhere near what I thought the market would do. It takes a long time to redesign backend controllers and I should have remembered that. This one did not happen as I expected.
5. We will see file system appliance vendors becoming application appliance vendors with specific highly tuned appliances embedded into the architecture.
This is happening all over the market from databases to data analysis to HPC file systems. This is the future trend, given the complexity of the problems and configuration and tuning of the systems.
6. We will see a significant reduction in the number of LTO tape library vendors and major cutbacks from some vendors. The reason is that LTO-6 only increased density to 2.5 TB, and enterprise vendors are already way beyond that in terms of density and performance. For large tape library configurations, LTO is more expensive than enterprise tape, given the cost of robot slots and number of tape drives needed to meet performance requirements. Though LTFS will help, it is not going to stop this trend.
I think that the jury is still out on this prediction, but a number of smaller players are seriously struggling to compete with the larger library vendors. I feel like I hear more about enterprise tape replacing LTO, but neither IBM nor Oracle will release numbers on drives or media.
7. We will see the T10 PI (host to disk data protection) standard finally become mainstream. There have been a few technologies that have caused this to lag, but I think we will finally get everything in place. If you are a regular reader, you know that I have made this prediction before and been wrong. We really need this technology.
Finally it is here, from the big players and some of the smaller ones also. This is good for our industry.
8. More consolidation in the HPC community will impact us all. Last year Intel purchased the QLogic Infiniband business, Cray’s network interconnect group, and WhamCloud, the Lustre file system company. Intel has clearly started a trend, and other vendors will have to respond.
At least at the time of this publication this has not happened. I still think it will happen, but it clearly did not happen as soon as I thought it would.
9. A vendor will demonstrate new non-volatile memory technology that could be used for things like database index tables. I am not saying who will demonstrate it or what the technology will be, but I believe we will see something next year.
There was an interesting article by Paul Shread on Nantero. Though according to the article we are not there yet, Paul made it sound like we are very close. I am going to put this one in the correct category, in terms of predictions.
My tally on predictions from last year: seven correct, three incorrect.
10. Finally, this last item is more of a hope than a prediction, but I will make it anyway. The industry really needs more than POSIX (open/fopen, read/fread,write/fwrite) and more than simple REST put/get interfaces for data in the future. Neither has the richness to address the myriad of polices that are needed in our future world. I predict that there will finally be some honest discussion about this among the customers that need it and the vendors that could create it. Maybe this should be my request to Santa. I have tried to encourage this discussion for years and I have gotten no traction.
As expected nothing is really being done in this area and I got no support from anyone to make changes. As said this 1 was a hope and not scored.
Looking back, I did far better than in previous years – but I still have a ways to go to achieve perfection. So what about this year? Here are 10 trends that I think are worth watching:
Looking Ahead: 2014 Predictions
The first few predictions are going to be a continuation of what I forecast last year.
Prediction 1: The SSD consolidation juggernaut is not over yet. In 2014 we will continue to see more companies declaring bankruptcy and/or being purchased. OCZ is a perfect example of this trend. The market is consolidating in different ways than how it proceeded in the disk drive market. The big NAND players are already starting to flex their collective muscles.
Prediction 2: PCIe 4.0 will be delayed again from the current date of late 2015 to some time even later. This will have a major impact on vendors that make peripherals and external connectivity such as NICs, HCAs and/or HBAs.
Prediction 3: The market for fibre channel and the percentage that fibre channel has for market share will continue its dramatic drop. The costs of SAS connectivity at 12 Gbit/sec and 40 GbE and FCoE will continue to reduce the marketshare held by fibre channel.
Prediction 4: Multiple vendors will be producing big data analysis appliances for a variety of applications specific requirements from point of sale to image change detection.
Prediction 5: We have a few entries into the REST appliance market for data access and archive, but next year the market will grow dramatically with at least three new vendors and new technologies.
Prediction 6: 2014 will finally be the year that someone does something about storage security and the management of who can access what, when. Too many people have had access to information they should not have had access to, and this needs to be logged and prevented.
Prediction 7: Enterprise tape will continue to far exceed the density of LTO. LTO density is seriously lagging and, if by 2015 this does not change, and IBM and Oracle continue on their enterprise tape density paths, this does not bode well for LTO.
Prediction 8: Parallel file systems will continue to take marketshare away from NAS appliance vendors because in large configuration they provide much more scalability. What is lacking at this point are some of the enterprise features, such as snapshots and remote replication. These features will be added and this will impact the upper end of the NAS market space.
Prediction 9: We will start to see the end of SATA usage in the enterprise and the industry will finally realize that SAS is an enterprise connectivity technology and SATA is not. With 12 Gbit/sec SAS already arrived and 8 Gbit/sec SATA still in committee, this should be obvious but it clearly has not been.
Prediction 10: Last but certainly not least: all major vendors will be using declustered RAID by the end of 2014, given the size of disk drives expected to be available.
Have a happy, safe holiday season and a prosperous 2014!
Photo courtesy of Shutterstock.