The I3en family is a class of instances clearly targeted at storage-intensive workloads, but CPUs are still needed to process that data. The largest of the I3en, i3en.24xlarge, ships with 48 cores of the Xeon Platinum 8175M, clocking at 2.50GHz.
At Scylla Summit 2018, Yahoo! Japan’s engineers Takahiro Iwase and Murukesh Mohanan took to the stage to describe their reasoning for testing Scylla, a challenger, against Apache Cassandra, their in-house long-time NoSQL favorite. With Cassandra, as Takahiro said, “We have problems. Lots of problems.”
“And now for our main event! Ladies and gentlemen, in this corner, weighing in at 34% of the cloud infrastructure market, the reigning champion and leader of the public cloud…. Amazon!” Amazon has unparalleled expertise at maximizing scalability and availability for a vast array of customers using a plethora of software products. While Amazon offers software products like DynamoDB, it’s database-as-a-service is only one of their many offerings. “In the other corner is today’s challenger — young, lightning quick and boasting low-level Big Data expertise… ScyllaDB!” Unlike Amazon, our company focuses exclusively on creating the best database for distributed data […]
Scylla 2.3 was just recently released, and you can read more about that here. Aside from the many interesting feature developments like improved support for materialized views and hardware enablement like native support for AWS i3.metal baremetal instance, Scylla 2.3 also delivers even more performance improvements on top of our already industry-leading performance. Most of the performance improvements center around three pillars: Improved CPU scheduling, with more work being tagged and isolated The result of a diligent search for latency-inducing events, known as reactor stalls, particularly in the Scylla cache and in the process of writing SSTables A new, redesigned […]
Benchmarking is no easy task, especially when comparing databases with different “engines” under the hood. You want your benchmark to be fair, to run each database on its optimal setup and hardware, and to keep the comparison as apples-to-apples as possible. (For more on this topic, see our webinar on the “Do’s and Don’ts of Benchmarking Databases.”) We kept this in mind when conducting this Scylla versus Cassandra benchmark, which compares Scylla and Cassandra on AWS EC2, using cassandra-stress as the load generator. Most benchmarks compare different software stacks on the same hardware and try to max out the throughput. […]
A fast in-memory database provides benefits that we all can appreciate such as optimal latency and throughput for workloads. What if you could utilize extremely fast NVMe drives to have similar latency and throughput results? The scope of this blog post is to examine the outcomes of using an in-memory like database combined with NVMe drives to resolve cold-cache and data persistence challenges. In this experiment, various testing scenarios were done with Scylla and Intel® Optane™ SSD DC P4800X drives with a goal of providing a solution with the performance of an in-memory like database without compromises on throughput, latency, […]
Visit us at AWS Summit New York on August 14th: Demos, fun games, and cool swag! When: August 14 7:30 AM – 7:00 PM Where: Javits Center Register to attend Going to AWS Summit New York? Be sure to stop by and visit ScyllaDB at booth #214 to see a demo, pick up a t-shirt, and play our fun game to win cool swag, including Scylla the Sea Monster.
By default, Scylla SSTables will be compressed when they are written to disk. As mandated by the file format, data is compressed in chunks of a certain size – 4kB if not explicitly set. The size of the chunk is one of the parameters for the compression property to be set at table creation. Chunk-based compression presents trade-offs that users may not be aware of. In this post, I will try to explore what those trade-offs are and how to set them correctly for maximum benefit. As trade-offs imply different results for different loads, we will focus on single-partition read […]
A database like Scylla can be limited by the network, disk I/O or the processor. Which one it is often dynamic and depends on both the hardware configuration and the workload. The only way of dealing with that is to attempt to achieve good throughput and low latency regardless of what is the bottleneck. There are many things that can be done in each of these cases that range from high-level changes in the algorithms to very low-level tweaks. In this post, I am going to take a closer look at fairly recent changes to Scylla which improved the performance […]
The Scylla team has reached a major milestone this week. We announced the release of version 1.0 and pronounced Scylla to be production ready. However, some early adopters have been using Scylla in production for the past 3 months. Recently a ScyllaDB customer migrated a 30-node Cassandra cluster to Scylla across 3 geographical data centers. Scylla is being evaluated by Fortune 500 companies in the e-commerce, finance and mobile industries.