AWS re:Invent has already come and gone, but its impact is still reverberating through the cloud computing industry. Since a sizable portion of Sisense clients use some kind of AWS product, the developments discussed at re:Invent have huge repercussions for them and for us.

Luckily, Sisense’s culture of innovation means that everything coming down the pike is something we’re already working on or are ready to tackle. The three technologies we’ll discuss in this post are AWS Lamdba (serverless computing platform from AWS), EKS (Amazon’s AWS-managed Kubernetes solution), and AWS storage solutions (EFS, Lustere, and EBS using high-performance SSD).

Serverless Computing

First off, let’s talk about serverless computing. As revealed at re:Invent, Amazon’s AWS Lambda and API Gateway are now able to run code without provisioning servers. The system already supported a variety of languages (Java, Go, PowerShell, just to name a few), but Ruby support was also announced at re:Invent. Plus, the Lambda Runtime API can now handle any other languages that users choose, increasing the flexibility of this versatile system. Serverless computing saves users money since they’re not paying when their servers are not running. It also increases security, as the code runs on AWS’s operating system and AWS can bounce the code from one server to another, allowing them to patch and update the operating system frequently. It also gives users increased elasticity, as AWS automatically scales out serverless services according to usage.

Analytics With Live Data From Redshift Databases: Answers to Your Most Burning Questions:

Get the free whitepaper

This is a game-changer not just for Sisense, but for any SaaS solution. When considering the architecture of a SaaS system, teams need to determine which services should be implemented using serverless technology. Sisense is actually uniquely well-suited to this change as most of the solution can be handled this way; the entire control plane and front end could all be serverless. Even query services and connectors could be good candidates for serverless computing. For clients, this can also drastically reduce the cost of hosting.

EKS: AWS-Managed Kubernetes

Another feature discussed at re:Invent that could reduce hosting costs is EKS, Amazon’s AWS-managed Kubernetes solution. EKS aligns with our new architecture: using Linux with a Kubernetes deployment. This means that we will be able to deploy our Kubernetes cluster in AWS environment without doing any adaptations. The big opportunity for Sisense clients here is the ability to manage ElastiCube instances efficiently in a cloud environment. EKS makes it easy to manage this sophisticated setup and helps save the client on resource costs. Additionally, the administrator can scale up the number of nodes by allocating more EC2 instances to the cluster.

AWS Storage Solutions

Lastly, we want to talk about the storage options Sisense can use in Amazon cloud. The simplest storage solution for the ElastiCube is EFS file systems. This system provides a fast network and reliable storage and sharing of ElastiCube Big Data files without the need to create and manage multiple replicas of the files. Additionally, Sisense users can benefit from storing those files on EBS volumes with EBS Provisioned IOPS SSD (io1), which provides some of the best performance available today.

And the latest storage option announced during AWS re:Invent is FSx: for Lustre, a fully-managed file system that is optimized for computing-intensive workloads. Amazon FSx can process massive data sets at up to hundreds of gigabytes per second of throughput, millions of IOPS, and sub-millisecond latencies. It’s seamlessly integrated with Amazon S3, so offline ElastiCubes can be stored on S3 for low-cost data storage.

Until Next Year

This is all extremely cost-effective. Users get cheap and performant non-replicated storage for processing data. With so many storage options available, a good SaaS architecture should empower users to employ the best storage solution for each ElastiCube instance according to their preferences or usage analysis.

So, there you have our breakdown of some reveals and discussions from re:Invent. If you couldn’t make it out or just didn’t have time to absorb the wealth of information, now you’re one step closer to grasping the implications of AWS Lamdba, EKS, and the new AWS storage solutions, both for your data and analytics needs and for Sisense.

Analytics With Live Data From Redshift Databases: Answers to Your Most Burning Questions:

Get the free whitepaper
Tags: |