Personally, at ‘OpenStack Summit Austin 2016,’ there were some events remained in my memory, but an inflection point in OpenStack’s future journey was seen.

Starting from the personal part, I did something memorable in Austin, in which it has been four years since I visited, but further, I felt that OpenStack’s status in the industry got off the ground to some degree as the atmosphere is changed in some ways when compared to the last Tokyo summit.

If ‘network’ was the message that strongly remained in the memory in the last summit, I suppose the message in this year is a capability as ‘integrated engine.’ The area of the capability as an integrated engine that accommodates all of hybrid infra environments is extended by meeting software defined network and storage. The businesses participated the marketplace were also interesting. It was impressive that, of course, from the server and cloud related businesses, EMC, Oracle to Cisco brought out integrations with OpenStack.

The importance of a new version ‘Mitaka’ and the future direction or ‘cooperation’ part, which was introduced in this summit, with other communities gives quite different feelings. Personally, in the ‘cooperation’ part, I strongly felt a change of the current situation in different attitudes compared to the last ones. In the survey result conducted on users in this year, the encouraging reaction was continued at more positive flow of reaction than the past results, but it seems that unstable factors on the backside are still remained.


The survey result in April, 2016, which OpenStack foundation presented is provided in the website, and I could conveniently see in a form of pamphlet in the pressroom of this summit. If personally choosing a positive result, I want to point out the result that the use of production level increased a lot and the cases in environment of a large scale are secured at the same time. It means that OpenStack is technologically already in a stable level in the market, over the start of a certain generation.

Among these result, in detailed users’ opinion part, we can see light and shadow that OpenStack has. Firstly, as positive parts, the stability that is improved after Kilo release, functional improvement as much as it can be used in the enterprise environment, and the aspect of open source and removing vendor dependency are pointed out. In technical aspect, the level of individual projects included in OpenStack is now on high-level tracks.

However, there is a feeling that a comment toward the problem, which should be solved, seems bigger. That integrating components is complicated, the lack of synergy among components and the excessive fragmentation are firstly pointed out. In fact, in community-based project that cannot control both detail projects and developers, this is often shown when the whole project of OpenStack is too large and the developers, who are concentrating on technology, excessively suit their tastes.


At the interview of the last summit, I think that the two questions that I asked touched a raw nerve. One of them was ‘what do you think about that users’ fatigue might get higher since 6 month release is too short?.’ At that time, the answer was ‘There is no problem because everyone is following well.’ In fact, in this survey, there is a result that the number of movements to a new release is increased, but it does not seem to be like that when looking at users’ opinions.

A fatigue about the release of the six months is shown in three parts; difficulty in distribution, non-persistence, and instability. A fast update means a quick end of support. The release support of OpenStack community is not that long, and at this point of time, security support of Kilo release, which is used the most commonly, is ended. Of course, this is a case when using community release by oneself, and if you want to get a long-term support, a commercial OpenStack solution will be essential.

Since the opinions about frequent releases and quick ends of support are reported in this paper, it is reasonable to say that many users are feeling the pain. For solving the problem, it seems that OpenStack needs to implement thoroughly standardized and modularized environment, by considering rolling upgrade, and start endless upgrade-journey, or to find methods for long-term support in many ways. Of course, it might be a symbiotic strategy that the foundation more focuses on development and the commercial solution takes the support part.


The other question in the last summit was ‘how will you create cooperation among projects from foundation?,’ and, at that time, the answer was ‘it has to be done among users and projects on their own, and there is no plan at foundation level.’ However, in this summit, the movements at foundation level, such as removing session category related to OPNFV, was shown, but I think that this might be a necessary part to empower ‘integrated engine.’ Also, I think that the movement in network part might have changed the whole direction.

Personally, I think that the movement to NFV is suggested into the direction of the ‘cooperation’ by meeting with OpenStack. As the cases that integrate and implement completely different infra regions in the same x86 based hardware and virtualization based environment come out, it seems that they moved smartly about the possibility of integrating across the software defined infra. The Superuser award in the summit selects a model case of ‘trend,’ and AT&T’s case was ‘integrated infra’ of computing and network.

It seems that the implementation of integrated engine through OpenStack reached the level of drawing a big picture by cooperation among foundations rather than just interlocking functions among individual projects. When thinking about the last answer, I somewhat grudge but also think that responding like this is just as well as the world changes very fast. Especially, by seeing that it expresses the cooperation into ‘an essential choice for survival,’ I rather think that I touched a raw nerve at that time.


Although several clear shadows exist, in this summit, I could clearly see that OpenStack’s spirits are getting higher. Especially, it feels strange to see that the businesses, who would be up in arms against the flow to OpenStack and put up their viable challengers in the past, try to get on the flow to OpenStack. Of course, each other’s strategies are different in this big flow, but it seems that the existence of OpenStack got so big and wide that the movements of the strategies cannot even confuse it.

As a meaning of open source is extending, I think that the change throughout the industry made a large contribution toward making strategy aspect of OpenStack’s positions as an ‘integrated engine’ and a ‘practical standard.’ OpenStack is already in the standard position of practical industry that subjects, who want to use it, widely participate, create and remove overlapping efforts with each other. The reason why this atmosphere is becoming stronger is that ‘support service’ business model, which can make differentiations even in a project, is settled.

Now, it seems that OpenStack met an important turn-off in the entire journey. Technologies that each component has are somewhat on the ranks, and many positive signals, such as that the practical effectiveness aspect began to be proven by the actual cases, that applicable range became wider, and that much more players are participating, are shown. Moreover, as the plan of the next one year is the period when ‘overcoming shortcomings,’ which was put off so far, becomes regularized, it is expected that projects and movements of market for this period will shake throughout the IT industry.

Copyright © acrofan All Right Reserved