Shulou(Shulou.com)06/01 Report--

This article will explain in detail what the new functions of EMC VNX2 are, and the content of the article is of high quality, so the editor will share it for you as a reference. I hope you will have a certain understanding of the relevant knowledge after reading this article.

Because the target promoted by VNX2 is very high, it is bound to form a competitive relationship for entry-level products of high-end storage. Therefore, I would also like to know more details before the release of EMC. After searching the Internet, I found a lot of information. It seems that the security work of EMC is still much worse than that of Apple. O (∩ _ ∩) O ha!

First, let's take a look at the specifications:

What I am more concerned about is the highest model VNX8000, which supports only 1500 hard drives, which is OK. How can the original propaganda support 16PB? How did you get it. Today, I carefully studied it and found that the VG8 NAS gateway was originally used. After that, a maximum of 4 sets of VNX can be connected and a maximum of 4000 disks can be supported. Therefore, according to the calculation of 4TB/ disk, 16PB can be supported. EMC, you are so fooling that you didn't almost scare me to death. If so, HDS VSP claims to support 255PB external storage, this extensibility is NB. Alas, today's manufacturers dare to brag about anything in order to attract attention. Similar to the shocking cosmetics advertisement published by Southern Weekly a few days ago, we ordinary people have been fooled.

Let's take a look at what are the new features and features:

The film found by a foreign website clearly lists the new features of VNX2. I found that I was cheated again. This 1 million IOPS is not only a SPC-1, but also a read-only performance indicator. The backend must be using SSD, and there must be FAST CACHE acceleration to achieve this performance. I even suspect that the host SSD card acceleration with XtremSW can achieve this performance. Therefore, although the index is high, it is of little significance for reference.

Other updates, in fact, the bigger touch for me is that the controller supports real dual living. This is a high-end storage technology, and HDS first brought this technology to the middle end. Now the middle end of HP 3PAR should also be this symmetrical dual-live architecture. It seems that EMC can not survive, but also follow the trend, launched a symmetrical double live in the middle. However, we can see that EMC does not seem to be fully supported, only traditional LUN is supported, and LUN for FAST VP should not be supported.

With regard to the three ways in which the controller works, I think many people are not clear. I would like to highlight it here. The following figure illustrates the history of the development of these three architectures:

At the beginning, EMC's CX series adopted the active / standby mode, that is, for a LUN, the host can only be accessed from one controller, and the other control is the backup relationship. When the primary controller fails, it switches to the backup controller. In the later stage of CX, EMC upgraded the system and began to support ALUA (Asymmetric Logical Unit Access) and extended to the VNX series. ALUA is an asymmetrical dual active mode, which is now used in most mid-range storage. That is to say, both controllers can receive the I / O of the host, but the LUN must belong to a controller at a certain time. Only the home controller can access the LUN. That is, the controller receives an access to the LUN to which it does not belong and must forward it to the controller that owns the LUN. However, the latest VNX2 adopts a symmetrical double-active approach to the ordinary LUN, that is, the LUN does not belong to a certain controller, and any controller can read and write to the LUN. The advantage is that the performance is better, and the LUN does not have to switch to the home after the controller failure, so the fault recovers faster. If it is in ALUA mode, the switching will have a great impact on the business.

Another major new feature of VNX2 I think is block re-deletion. At present, in the mainstream mid-range array manufacturers, data re-deletion is realized at the file level, including NETAPP. EMC VNX2 implements re-deletion at the block level, which should be the first mid-range storage product among mainstream manufacturers to support block re-deletion. Of course, VNX2 also supports data compression, but it cannot be shared with block re-deletion.

Another useful feature for users is that the migration granularity of FAST VP has changed from 1G to 256m. Although the granularity is still a little larger than that of other competitors, it is much better.

Generally speaking, apart from adopting the latest hardware technology, software re-optimization and performance improvement, VNX2 has not made a big change to the architecture of VNX. In particular, there is no feeling of progress in the following areas:

Unified storage architecture. VNX2 still uses the VNX file and block engine separation method, does not really achieve the integration of files and blocks, and even the hardware and software of the NAS engine is basically not updated, which is really a bit unreasonable. It is said that EMC would have made a truly integrated version of VNX2, but the R & D team could not do it, so the head of development was rumored to have been replaced.

High-speed interface. With PCIe 3.0, it should be possible to introduce a higher-speed interface. In the current release, we can't even see 16G FC interface and 40G Ethernet interface. 16G FC is now supported by many manufacturers. I read an article on DOSTOR a few days ago and found that the 32G FC will be on a large scale next year. However, it is a pity that there is no new interface type for new VNX2 products.

Perhaps for these reasons, EMC did not adopt a new series name for VNX2. It still uses VNX, but the model is higher than the original one. This also shows that the update is limited. Compared with the CX series, VNX changed greatly in terms of modular card insertion and back-end SAS, coupled with the packaging for unified storage, so a new series name was used. Now this VNX2, the architecture remains the same, the interface card remains the same, and there are too few changes, so it can only be regarded as a small upgrade.

This leaves a lot of room for domestic manufacturers to catch up again. The architecture uses the latest CPU and chip, with SSD, the performance should not lag behind VNX2. At the same time, unified storage with converged architecture is launched as soon as possible to support high-speed interfaces, so it is not impossible to partially surpass VNX2.

About what the new features of EMC VNX2 are shared here, I hope that the above content can be of some help to you, can learn more knowledge. If you think the article is good, you can share it for more people to see.

Welcome to subscribe "Shulou Technology Information " to get latest news, interesting things and hot topics in the IT industry, and controls the hottest and latest Internet news, technology news and IT industry trends.