Thursday, September 19, 2024
Home » 3 big takeaways from Scality’s Service Provider Summit

3 big takeaways from Scality’s Service Provider Summit

Scality’s 6th Service Provider Summit brought cloud service providers together to share their experience with Scality’s products across  their own infrastructure and applications.

Many common open source components were mentioned. As expected, every customer has heterogeneous environments, using a wide variety of packages like Cassandra (for metadata storage), Dovecot, OpenXchange, and Zimbra. OpenStack was also mentioned a few times, along with Elasticsearch and more.

Scality’s limited presentation presence included a keynote by CEO Jerome Lecat, who talked about digital transformation and how it affects Scality users. On day two, Scality CTO Giorgio Regni shared the vision for the company’s next-generation products and Benjamin Morge,  Scality’s Director of Global Support presented the work of his team. The rest of the time was spent offering technologists a safe space to talk about their specific operations, sharing the details of their installations, their pain points, and many exciting innovations with their peers. Here are some of the big takeaways from the event.

  1. Ideas flow faster when users talk and vendors just listen. The exchange of knowledge between Scality professional services and RING users advances the whole community much faster than any 1:1 training can. The summit agenda allowed plenty of time for each speaker to answer questions and clarify any doubts. The time was well-managed, the conversations fluid.
  2. To collect honest feedback from customers, leave the room. The final feedback session was done in two stages: first the customers met without Scality employees present. We all got up and left so that customers and partners could sit by themselves and take notes in three areas: what Scality does well, what should improve, and which features are missing in the product. After they worked for about 45 minutes, they called us back in to review the notes. It was brutal honesty in action: the group didn’t hold back their punches or praise. There weren’t many surprises, but rather confirmation of the progress we made from the previous year’s feedback along with new areas that require attention.
  3. Messaging platforms as walled gardens are a problem for everybody. OpenXchange shared details about the Chat Over IMAP (COI) initiative. With a strong technical audience, it was a good test for the pitch. The room was in agreement that proprietary messaging platforms like Facebook and WeChat are a problem and welcomed the fresh approach. Ideas for improvements kept flowing during the break, too.

Another technical conversation revealed that data migration is a fact of life for any RING (and storage) user. A new customer told me over coffee: “I’m so glad I heard about the stories of growth now, so when I have to increase my capacity I know what to expect.”

Another story from the scale-out trenches highlighted the role of networks in RING clusters. Network disturbances will happen and they can destabilize the RING. Every hop in a network introduces possible failure points. Latency is less of an issue but networks going down will lead to misplaced objects. The recommendation: “Plan for chaos! ‘Wargames’ style, shut down the network and see what happens.”

On the topic of growth, one customer faced an unexpected growth of storage requirements. Their predictions were based on a linear consumption until, suddenly, the curve changed angle.  They learned that hardware orders never arrive on time and in perfect condition. There will always be a RAM stick or a RAID missing. Or some of the hardware will fail for mysterious reasons once mounted on a rack. Lesson: “It’s better to have as much hardware available upfront to speed up upgrades and expansions.”

On the positive side, one customer reported that fine-tuning RING can greatly exceed initial estimations. They saw in production about 1.7 times the OOPS performance than they expected after the sizing exercise.

Finally, reputation counts. Gartner Magic Quadrant is a great tool to identify the leaders but as one customer put it, “you need to look a little deeper at your specific use case before picking your leader.” Also, many pointed out that they look at what their peers are buying and Scality’s reputation is great in the service provider community. What an honor to have such great references.

Leave a Comment

* By using this form you agree with the storage and handling of your data by this website.

About Us

Solved is a digital magazine exploring the latest innovations in Cloud Data Management and other topics related to Scality.

Editors' Picks

Newsletter

Challenges solved, insights delivered, straight to your inbox.

Receive hand-picked articles, case studies, and expert opinions. Keep up with industry innovations and get actionable insights to optimize your strategy.

All Right Reserved. Designed by Scality.com