Download as PDF, PPTX
Uploaded byScyllaDB
Comparing Apache Cassandra 4.0, 3.0, and ScyllaDB
The document presents a comparison between Apache Cassandra versions 3 and 4, and ScyllaDB, focusing on performance improvements and key features. It highlights Cassandra 4's enhancements such as support for JDK 11, increased speed and scalability, and better compression settings, while showcasing Scylla's superior performance in latency, throughput, and node management. The conclusions suggest that Scylla not only outperforms Cassandra but also offers long-term savings and additional features that could outweigh the challenges of migration.
In this document
Powered by AI
Comparison of Apache Cassandra versions (3 & 4) and ScyllaDB's capabilities as a drop-in replacement.
Comparison of Apache Cassandra versions (3 & 4) and ScyllaDB's capabilities as a drop-in replacement.
Introduction of presenters, Karol Baryła and Piotr Grabowski, highlighting their backgrounds and roles at ScyllaDB.
Outline of topics covered: Testing methodology, comparisons between Cassandra versions, and conclusions.
ScyllaDB as an open-source, high-performance database, designed to compete with Cassandra and DynamoDB.
Details on the improvements in Cassandra 4.0 including speed, scalability, new features, and release cycle.
Methodology for performance tests including latency measurements, throughput evaluations, and node management.
Comparison setup for tests: hardware specifications for clusters using Cassandra 3, Cassandra 4, and Scylla.
Performance testing data based on Gaussian distributions with mixed, disk-intensive, and memory-intensive workloads.
Tests on cluster node management operations such as adding nodes and replacing nodes.
Discussion on Java support in Cassandra 3 and 4, and how Java 11's introduction of ZGC enhances performance.
Poll regarding the amount of data participants manage in their transactional database systems.
Performance testing data comparing Scylla to both Cassandra 4 and 3 across various workloads.
Similar node management benchmark tests performed in Scylla comparing various operations.
Cost and performance comparisons of Scylla on 4 nodes versus Cassandra 4 on 40 nodes.
Convergence of test results from disk-intensive, mixed workloads, and cluster optimizations.
Conclusions from findings indicating the significant performance improvements of Scylla over Cassandra.
Factors to consider for upgrading to Scylla from Cassandra, emphasizing performance and cost benefits.
Call to action for downloading Scylla and learning more through provided links.
Q&A session details, contact information for presenters, and acknowledgment of the audience.
More Related Content
![[OracleCodeTokyo2019] Kubernetesで実現する運用自動化の新しいアプローチとは](https://cdn.slidesharecdn.com/ss_thumbnails/oraclecodetokyo19b-2-2k8soperatorkshigeruv1-190522024836-thumbnail.jpg?width=600ounds&width=560&fit=bounds)
Similar to Comparing Apache Cassandra 4.0, 3.0, and ScyllaDB
![UiPath Automation Suite Installation (Hands-On) [2/3]](https://cdn.slidesharecdn.com/ss_thumbnails/automationsuitecommunitysession2-251015095633-a6d862f1-thumbnail.jpg?width=600ounds&width=560&fit=bounds)
Comparing Apache Cassandra 4.0, 3.0, and ScyllaDB
- 1. Comparing Apache Cassandra 4,3, and Scylla Open Source
- 3. Comparing Apache Cassandra 4,3, and Scylla Open Source
- 4. Presenters 4 Karol Baryła Karol isa Junior Software Engineer at ScyllaDB. He often participates in security CTF competitions as a member of team "Armia Prezesa" where he solves web security and reverse engineering tasks. He is currently pursuing an MSc in Computer Science at the University of Warsaw. Piotr Grabowski Piotr is a software engineer working at ScyllaDB. From a young age, he participated in many competitive programming contests. Piotr holds a BSc in Computer Science from the University of Warsaw and is now pursuing an MSc. For the past year, he worked on Kafka connectors and Scylla Java Driver.
- 5. 5 1. Testing methodology 2.Cassandra 3 vs 4 3. Scylla vs Cassandra 4 4. Conclusions Agenda
- 6. 6 + The Real-TimeBig Data Database + Drop-in replacement for Apache Cassandra and Amazon DynamoDB + Outstanding performance & low tail latency + Open Source, Enterprise and Cloud options + Founded by the creators of KVM hypervisor + HQs: Palo Alto, CA, USA; Herzelia, Israel; Warsaw, Poland About ScyllaDB
- 7. At July 27th,2021 Cassandra team released version 4.0 - 6 years after the release of version 3.0. Let’s see how much Cassandra improved during those 6 years, and how well it holds against Scylla 4.4 now. 7
- 8. 8 1. Increased speedand scalability a. Zero Copy Streaming streaming data up to 5x faster b. Up to 25% faster throughput on reads and writes 2. Support for JDK 11 3. New configuration settings, better security and observability 4. Better compression settings (support for Zstd) 5. A shift to a 12-month release cycle Cassandra 4.0 new features
- 9.
- 10.
- 11. 11 1. Latency atdifferent throughputs a. Gaussian distribution b. Disk-intensive distribution c. Memory-intensive distribution 2. Adding a single new node 3. Doubling cluster size 4. Replacing node Benchmarked operations
- 12. 12 + 3 vs3: + Cluster nodes: 3x i3.4xlarge (16vCPU, 122GiB RAM, up to 10Gbps network, 2x1.9TB NVMe) + Loader nodes: 3x c5n.9xlarge (36vCPU, 96GiB RAM, up to 50Gbps network) + 4 vs 40: + Scylla cluster: 4x i3.metal (72vCPU, 512GiB RAM, up to 25Gbps network, 8x1.9TB NVMe) + Cassandra cluster: 40x i3.4xlarge (16vCPU, 122GiB RAM, up to 10Gbps network, 2x1.9TB NVMe) + Loader nodes: 15x c5n.9xlarge (36vCPU, 96GiB RAM, up to 50Gbps network) + Java version: JDK 16 (Cassandra 4.0), JDK 8 (Cassandra 3.11) Benchmarks setup - 3vs3 and 4vs40
- 13. Cassandra 3 vs4 13
- 14. 14 Gaussian distribution, mixedworkload
- 15. 15 Gaussian distribution, mixedworkload
- 16.
- 17.
- 18.
- 19.
- 20.
- 21. 21 Adding a singlenew node
- 22.
- 23.
- 24. 24 + Cassandra 3officially supports only Java 8 + Cassandra 4 officially supports Java 8 and Java 11 + Java 11 introduced ZGC - as an experimental feature + ZGC is considered production ready from Java 15 + We used Java 16 in benchmarks in order to utilize full power of ZGC + ZGC has extremely short pauses, which reduces Cassandra’s tail latencies. What causes latency improvements?
- 25. 25 How much datado you have under management in your own transactional database systems? + <1 terabyte + 1 to 50 terabytes + 50-100 terabytes + >100 terabytes Quick Poll
- 26. Scylla vs Cassandra4 vs Cassandra 3 26
- 27. 27 Gaussian distribution, mixedworkload
- 28. 28 Gaussian distribution, mixedworkload
- 29.
- 30.
- 31.
- 32.
- 33.
- 34. 34 Adding a singlenew node
- 35.
- 36.
- 37.
- 38. Scylla on 4nodes vs Cassandra 4 on 40 nodes 38
- 39. 39 + 3 vs3: + Cluster nodes: 3x i3.4xlarge (16vCPU, 122GiB RAM, up to 10Gbps network, 2x1.9TB NVMe) + Loader nodes: 3x c5n.9xlarge (36vCPU, 96GiB RAM, up to 50Gbps network) + 4 vs 40: + Scylla cluster: 4x i3.metal (72vCPU, 512GiB RAM, up to 25Gbps network, 8x1.9TB NVMe) + Cassandra cluster: 40x i3.4xlarge (16vCPU, 122GiB RAM, up to 10Gbps network, 2x1.9TB NVMe) + Loader nodes: 15x c5n.9xlarge (36vCPU, 96GiB RAM, up to 50Gbps network) + Java version: JDK 16 (Cassandra 4.0), JDK 8 (Cassandra 3.11) Benchmarks setup - 3vs3 and 4vs40
- 40. 40 Cost of theclusters
- 41. 41 Disk intensive, mixedworkload
- 42. 42 Disk intensive, mixedworkload
- 43.
- 44.
- 45.
- 46. 46 Summary of results +Cassandra 4 has much better tail latencies than Cassandra 3. + Scylla performs 3-4 times better than Cassandra when adding/replacing nodes. + Scylla adds 25% capacity to a 40 TB optimized cluster 11x faster than Cassandra 4.0. + Scylla performs major compaction 32x faster than Cassandra 4.0. + Scylla has 2x-5x better throughput than Cassandra 4.0 on the same 3-node cluster + Scylla has 3x-8x better throughput than Cassandra 4.0 on the same 3-node cluster while P99 <10ms + A 40 TB cluster is 2.5x cheaper with Scylla while providing 42% more throughput under P99 latency of 10 ms
- 47. 47 Should I upgradeto Scylla or Cassandra 4? 1. Upgrading is hard, so why not upgrade to Scylla right away? a. Upgrading is problematic anyway - you should make backups, you risk downtime. b. Migrating from Cassandra to Scylla is a bit more involving - but the benefits are worth it. 2. Upgrading to Scylla will save you the money in the long run. 3. Scylla offers better performance and lower latencies compared to Cassandra 4. 4. Scylla offers exciting new features: a. Scylla CDC b. Kubernetes support with Scylla Operator c. Scylla Cloud
- 48. Download Scylla OpenSource: scylladb.com/download Learn more https://university.scylladb.com/ Experience Scylla for Yourself 48
- 49. Q&A piotr.grabowski@scylladb.com Stay in touch Joinus at P99 CONF p99conf.io October 6-7, 2021 karol.baryla@scylladb.com
- 50. United States 545 FaberPlace Palo Alto, CA 94303 Israel 11 Galgalei Haplada Herzelia, Israel www.scylladb.com @scylladb Thank You!