Founded in 2004, QNAP Systems, Inc. designs and manufactures network-attached storage (NAS) and network video recorder (NVR) solutions for small, medium, and enterprise-level businesses. The company has received multiple industry awards for its NAS solution, which was listed by PCWorld as one of the 50 Best Tech Products of 2013. Headquartered in in New Taipei City, Taiwan, QNAP has offices in 10 countries to serve its international customer base.
By 2013, QNAP knew that it had to become more efficient at extracting insights from customer data. The company was already using AWS for its myQNAPcloud service. This service enabled customers to back up their local NAS data to Amazon Simple Storage Service (Amazon S3) and Amazon Glacier. QNAP collected event logs and usage statistics from the service, but wanted to make better use of the information.
“By running more responsive queries, we hoped to better understand how our customers were using QNAP products and harness these insights to make plans for our future development,” says Paul Chu, QNAP Cloud Services Director. In early 2014, QNAP’s web operation team began developing a new analytics platform to deliver the required customer insights. QNAP aimed to complete the project and have a functioning analytics platform within six months.
Why Amazon Web Services
“We initially considered using our own physical machines to support the new analytics platform. However, we quickly realized that a cloud-based service would be more efficient, both in terms of cost and operation,” says Chu. “We already knew that the AWS Cloud could support the rapid development and launch of new products and services, so it was always clear to us that we would use it to support our analytics platform.”
Building on AWS allowed QNAP to access resources as needed without affecting performance. “We were also reassured by the integration of strong security features into the AWS platform’s architecture,” says Chu. “We knew we could build on that to encrypt data and protect our network.”
Chu and his team were also impressed by the ability to connect different branch offices and data centers across the globe with consistent upload and download speeds using AWS. Most importantly, during QNAP’s four years of using AWS to support the myQNAPcloud service, Chu has found the company to be trustworthy, customer-focused, and quick. “AWS offers straightforward management interfaces and a comprehensive software development kit,” says Chu. “We’ve also been impressed by the company’s prestigious reputation and customer support.”
Working independently, QNAP spent six months implementing its new platform on AWS, which went live in March 2014. The company now runs its analytics platform in Amazon Virtual Private Cloud (Amazon VPC), allowing it to use security groups to control access to the platform, and Amazon Elastic Compute Cloud (Amazon EC2) for scalable compute capacity. With around 200 million log entries (equivalent to 10GB of log data) generated per day, QNAP uses between six and 10 Amazon EC2 instances for data analysis, scaling them according to demand. “We add more instances if we need to process extra data or extract more information,” says Chu. The company uses Amazon S3 for data storage, and Elastic Load Balancing for performance and stability.
QNAP also encourages its NAS customers to back up their data. Initially, the company used Amazon S3. Today, it uses Amazon Glacier, which—for QNAP’s purposes—is comparatively inexpensive. By using Amazon Glacier for its archiving needs, QNAP is able to provide its customers with a simple and cost-effective solution for backup, disaster recovery, and long-term archiving.
Customers can also access their data and files wherever they are by using myQNAPcloud, or take advantage of QNAP’s mobile apps to access and manage their devices using myQNAPcloud CloudLink.
Figure 1 shows the architecture for the analytics platform on the AWS Cloud.
Figure 1. QNAP Platform Architecture on AWS.
“Before building its analytics platform on AWS, QNAP had to search data logs manually and run data queries one at a time to generate statistics and business reports—an inefficient and error-prone process. “Depending on a query’s complexity, it could take hours or even days to see results,” says Chu. “Now that our analytics platform is in place on AWS, we can run previously time-consuming queries in a matter of minutes.”
QNAP can easily generate a wide range of business reports using the gathered data, providing QNAP with a degree of business insight it previously couldn’t achieve. “The platform provides us with a very detailed overview of user behavior and overall system health,” says Chu. “We can also forecast and prepare for any changes in customer usage patterns. AWS helps us to better understand how our customers are using the cloud services. This, in turn, allows us to make our services more stable and reliable.”
The success of the analytics platform has encouraged QNAP to look for other ways using AWS could help it achieve its business goals. For example, the company is planning to use Amazon Kinesis and Amazon Redshift in the future to provide real-time analytics capabilities. “We’ve been delighted with the practicality and performance of AWS,” says Chu. “Moving ahead, we’re excited about what else we might achieve using our AWS solution.”