We were researching database resiliency solutions, to be able to deploy software updates with zero downtime. Continuent Clustering has performed well, it is relatively easy to manage, and the support is excellent.

Having tried traditional ETL tools and slow data-scraping techniques that put a heavy load on operational systems, we found that we were unable to meet the low latency required by our business as well as capture a complete record of transactions. Hadoop has clearly changed the landscape of data management by providing a central data backbone hub that receives data from across the enterprise business; however, there is now a wide range of services that need to move transaction OLTP data efficiently into Hadoop. With Tungsten Replicator we can quickly and flexibly move data out of operational databases into Hadoop where we run analytics that answer important business questions on timelines matching the needs of our users.

We were looking for solutions to asynchronously monitor data from MySQL to Vertica and also have fully functional MySQL with cross data center replication, auto failover and load balancing capabilities. It’s always a sweet suite to have Continuent support. The engineers were very fast and assertive in dealing with any kinds of support tickets. Having a completely HA MySQL gives us a contemporary solution to focus more on building rigorous applications and having not to worry of database being down is the best.

We are a payment processor – so building a highly resilient platforms is a must. I was looking for something that would let us do multi-master replication between our two sites. We have to be able to switch processing between sites at a moment’s notice in the event of any system issue and come up at the other site. I wanted a commercial product because I didn’t want a homegrown solution we had to maintain. I needed my developers to be working on revenue producing projects – not maintaining or wasting time on a custom solution. I knew we would never be able to have the time to maintain something custom, and you are continually adding enhancements. I also really like the abstraction Continuent Clustering provided to the developers. Developers can treat the cluster as a single entity. They don’t need to manage where to read and write. And for our infrastructure team, having a slave promote to a master automatically if a master fails was nice additional resiliency. But mostly the solution let me build a site failover capability that works transparently to our clients. When there is a problem, we can re-route traffic immediately without having to worry about keeping data bases in sync. The issue for us isn’t performance. It is resiliency. This is critical to us delivering 99.999% availability to our customers. Our service offerings never go down (sound of me knocking on wood). Without Continuent Clustering, we couldn’t do what we do.

Continuent Tungsten is a really cool product. We can swap back-and-forth between our master and any number of slaves without losing a single query in the process. We can take a one offline, do updates, restart it. We can do schema changes to the slave and bring it up as a new master. So we can do a manual version of pt-online-schema-change, but it helps that by taking the system totally offline, a lot of schema changes for us go a lot faster than with pt-online-schema change, because it is not happening live on the server as it's being used for traffic. So lot's of cool things there.

I have nothing but great things to say about Continuent Tungsten. It's is an easy-to-use database replication and HA failover product that I have employed as the solution for our High-Stakes online assessment where 99.9% uptime is required. We also utilize it for our lower stakes platform that has 10 years worth of data. Its simpler to install and maintain than Linbit with Heartbeat, and the commands are intuitive and easy to use. The Continuent Tungsten documentation covers all the topics and the product itself makes server maintenance a breeze. The engineers at Continuent are fast to react to support cases, a much welcome departure from the Sun/Oracle help desk to which we were accustomed. And they have gone beyond the service agreement and helped us with other MySQL issues. I'd endorse Continuent Tungsten to anyone who needs full-time data availability and does not have team of experts at the ready.

Most important thing for us was the high-availability and the automatic fail-over features that were missing from the standard database features. With Continuent Tungsten, we can now truly say that our databases are clustered HA platforms. Daily database maintenance operations can be now done without causing long service breaks to the services. This is one BIG helping thing from DBA's point of view. Continuent Tungsten Smartscale out-of-the-box is a good concept that we utilize with our clusters. With new Continuent Tungsten support platform in use, we have got fairly quickly help to our problems, even during out-of-office times. This is how it should be and I hope this will stay like this in the future too.

We have Continuent Tungsten in all of our environments, (dev/qa/uat/prod) and we're using it for all of our web applications. We sell almost 1M used cars a year using Continuent Tungsten. I've been really impressed with the intelligence of both the software and the support we receive from Continuent. I have nothing but good things to say about your software and your company. I'd be happy to talk to any perspective customers at the Percona Live: MySQL Conference & Expo 2013 or take them to dinner while we're there. We currently have 100 MySQL servers running Continuent Tungsten and love the ease of use and management.

Continuent Clustering has allowed us to deploy clusters in the Amazon Cloud quickly, which is key for our rapidly growing company. In addition, Continuent Clustering is a key component in allowing us to meet our uptime SLA. Perhaps the most important item is the support we receive from Continuent, which I am pleased to say, responds quickly and resolves all issues (including self-inflicted!), and accepts feedback for future enhancements.

We have been using Continuent Tungsten products for over two years. Continuent Tungsten has taken a lot of the up and down issues away from my daily concerns. We use the replication to transmit MySQL cluster (running in AWS) data serving over 400 e-commerce stores into our Oracle data warehouse, which is living on another data center with Rackspace. The speed and consistency of the results is a big plus, but the biggest reason we stay with Continuent is their ability to get high level engineers and management involved if there is an outage or some catastrophic issues. Amazon outages are always a concern for a 24x7 shop like ours, but having this technology in our stack brings a lot of peace. We would like any future failures to fail over automatically and be able to sense what is the right course of action for us.

Millions of researchers, scholarly writers, students, and librarians use Thomson Reuters' EndNote Web to collect, organize and share their research. Cite While You Write in Microsoft Word easily enables users to cite references in their papers and create formatted bibliographies. Continuent Tungsten speeds up the development and performance of EndNote Web's dynamic, database-driven application and web services that must manage a complex set of workflows for millions of large, user writeable data sets.

Digital Spy is a recent addition to the Hearst Magazines UK Digital Department, which is dedicated to creating and implementing Hearst Magazines' digital strategy, and now has more than 25 websites and 15 mobile sites in the portfolio. High availability and data integrity are essential to our business. Continuent Tungsten is one of the key technologies, enabling us to deliver more than 150 million page impressions of the latest entertainment news and opinion to over 12 million unique users every month. With Continuent Tungsten, we get fast, predictable performance, which automatically manages the replication of data and workload over our two data centers.

MSS Operations has been using Continuent Tungsten for over a year and half in multiple global production deployments. In addition to eliminating single point of failure at the database layer, Continuent Tungsten has given us the ability to switch database nodes for normal operations and maintenance tasks as well as schema updates with minimal or no downtime. We have successfully upgraded to a newer MySQL version without a downtime or hiccup from an application perspective. Continuent support has been excellent and readily available during upgrades and to bounce off ideas and review standard operational procedure documentation.

Discovering Continuent was a real game changer for Stitcher in terms of our data redundancy, performance, scalability and up time. Using Continuent Tungsten and the Tungsten Replicator for our MySQL data, we were able to seamlessly migrate to AWS from our hosted datacenter with no downtime, have improved our SQL response time using read/write splitting, have nearly eliminated downtime caused by a DB problem, and are fully redundant with nodes distributed across AZs. Using Continuent Tungsten has simplified the task of working with terabytes of data by providing full backup, recovery, replication and failover support. I've also been very pleased with the Continuent Tungsten team. Whenever I have a question or issue in need of resolution, I've received prompt, expert and personalized service from them. The whole Continuent Tungsten environment has been one of the coolest technologies I've worked with.

Most of our MySQL database hosts are currently utilizing Continuent Tungsten replication. We mainly used fan-in for the migration of our legacy hosts to the newer hardware during 2012. That project was a success. As for the support, we just renewed our Continuent Tungsten subscription. We take advantage of that as much as we can when possible. It's always to have an extra pair of eyes when it comes to troubleshooting.

We use Continuent Tungsten for HA master/slave failover and do look forward to testing and implementing multi-master. It is in use for over 40-50 systems and has allowed us to quickly switch from master to slave in several situations where the motherboards failed and keep consistently running (to the application) with no interruption. We've had the opportunity to collaborate with Continuent on Tungsten quite a bit, which means it made integration into our environment for CyberArk password vaults and at a lower level with Puppet. As a result we can with confidence mass deploy our servers. We have great response time and collaboration on our issues and it is very nice to have that relationship with Continuent.

Millions of researchers, scholarly writers, students, and librarians use EndNote Web to collect, organize and share their research. Cite While You Write in Microsoft Word easily enables users to cite references in their papers and create formatted bibliographies. Continuent Clustering speeds up the development and performance of EndNote Web’s dynamic, database-driven application and web services that must manage a complex set of workflows for millions of large, user writeable data sets.

The only thing Marketo has more passion for than its product, is its customers. By leveraging the Continuent Clustering platform, we have increased our ability to provide higher levels of availability and uptime, including complex database upgrades and operations in production. Upon database failure, our earlier solution resulted too often in downtime, which is in excess of 30 minutes and impacting more than 150 customers at the time. With Continuent Clustering there is no down time, just predictable degradation of the performance until misbehaving servers will be up again. With Continuent Clustering we can make changes within our operation environment while being able to provide continuous operation during the regular maintenance cycles and also during larger upgrade process, being that for our own app, database or hardware. Continuent can help us at least in two ways to lower op-ex: enables more customers on each pod (database cluster) and allows us to use replication between our current hosting provider and another, more cost-effective hosting provider to augment pods or completely switch over to save significant money. These two combined will easily save us multiple times the investment we are making in Continuent Clustering software.

We were a MySQL centric shop. Our customer-facing site, as well as our backend systems, was written on top of MySQL. But with the acquisition of Zappos by Amazon, who is very centric on Oracle, we had a lot of requirements to start moving real-time data from MySQL 5.1 to Oracle 11g. We really like Continuent Tungsten because of its straightforward installation, monitoring of replication, quick turn around by Continuent team on new filter request, and helpful consulting service.

We use Continuent Tungsten to support our largest client. The client holds conferences around the world and needs high availability and scalability to maintain their registration systems. We utilize Continuent Tungsten in a traditional master/slave relationship with one master and two slaves. While we've had issues with some unique challenges regarding temporary tables, Continuent Tungsten has enabled us to maintain a very high uptime while putting minimal work on our staff. We're in the process of upgrading to the latest version and look forward to replacing some of our in-house built shell scripts with the wonders of tpm. The support has been wonderful. Whether it's simply talking through a setup issue with Jeff Mace or doing a screen share with Robert Hodges, we've always received timely and accurate information.