Tungsten Clustering vs. Amazon Aurora

MySQL and MariaDB Clustering Solutions Comparison

Learn why Tungsten Clustering is a better (complete, proven) and more cost-effective (software, 24/7 support and consulting all included) MySQL High Availability (HA), Disaster Recovery (DR) and Geographically-Distributed (Geo-Scale) solution compared to Amazon AWS Aurora for MySQL.

For any MySQL clustering to be a complete high-availability and disaster recovery solution, it requires three core components working in sync: connectivity (proxy), management (orchestration) and replication (local/global).

Tungsten Clustering

Tungsten Clustering includes three highly integrated and fully tested components Tungsten Proxy, Tungsten Manager and Tungsten Replicator working together to provide both local (individual clusters) and global (multiple clusters working together) data service for continuous 24/7 MySQL operations.

Connectivity Complete and Integrated (Tungsten Proxy)

Orchestration Integrated Business Rules (Tungsten Manager)

Replication Advanced Local and Global (Tungsten Replicator)

 Connectivity
 Orchestration
 Replication

Amazon Aurora

Amazon Aurora for MySQL is an attractive, yet somewhat expensive, ‘managed’ SaaS solution for DevOps and local MySQL HA needs. Adding disaster recovery will make it much more expensive without performance gain on the secondary site. Aurora does not offer a truly global multi-site solution.

Connectivity Direct (benefits from 3rd-party proxy or load balancer)

Orchestration Limited

Replication Distributed file system for cluster, asynchronous for AWS cross region

Amazon AWS Aurora for MySQL Comparison

Use Cases Supported

MySQL HA

Supported

MySQL DR

Supported

With Aurora Global Database

MySQL Zero Downtime Maintenance

Supported

Not Supported

MySQL Continuous Operations

Supported

With Aurora Global Database

MySQL Multi-Primary

Supported

Not Supported

MySQL Active/Passive Clusters

Supported

Not Supported

MySQL Active/Active Clusters

Supported

Not Supported

MySQL Geo-Scale Clusters

Supported

Not Supported

Features & Benefits

Continuous Operations

4.67

3.00

Zero Downtime Maintenance

Both database, operating system and hardware changes & upgrades

Requires maintenance window and downtime

High Availability

Automatic failover of primary

30 seconds downtime with DNS change

DR with Active/Passive Cluster

Automated site level switch with one failover command

Only with Aurora Global Database. Takes “a few minutes.”

DR with Active/Active Cluster

Only requires repointing the apps

Cross-region active/active not supported

DR Failback

Single command, fast

Cross-region failback requires reprovision

Backup & Data Recovery

Flexible backups using existing external tools

Managed, proprietary

Scalability/Read replicas

4.40

4.00

Scalability/Read replicas

Supported

Scalability/Physical location

Deploy anywhere, on-prem and in the cloud (also hybrid- and multi-cloud)

AWS regions only

Scalability/Instance size

Choose the instance size, minimal requirements

Supported

DB Node Auto-Scale

No auto-scale. Requires manual steps

Auto-scale supported with "serverless," reduced functionality

DB Cluster Scaling

Ideal for the SaaS apps (Pod Architecture)

Ideal for DevOps

Geographic Scaling

5.00

3.33

Scalability/Read replicas

Across regions, availability zones, and clouds

But only for read replicas and in AWS regions

Active/Active Geo-Clusters

Multi-site active/active supported and widely deployed

Local only

Active/Passive Geo-Clusters

Supported and widely deployed in cross-regions and cross-clouds, also on-premises

Supported (in AWS only)

Performance

4.75

3.50

Smart-Scale

Tungsten Proxy with automatic read/write split, no application changes

Not natively, requires 3rd party proxy

Load Balancing

Highly configurable with automatic read/write splitting

High performance but read/write splitting requires app awareness

Up to Date Read Replicas

Using asynchronous replication

Using distributed file system

Robust Database Performance

Tune your favorite database, asynchronous replication adds almost no performance penalty

Tweaks to standard MySQL for performance

Transparency

5.00

3.33

No Application Changes

Works out of box without app changes

Apps with advanced MySQL 8.0 and MariaDB features may not work

All MySQL Versions

Native MySQL: Community, MySQL Enterprise, MariaDB, Percona Server, all versions

"Compatible MySQL" based on MySQL 5.6 and 5.7, no MySQL 8.0 or MariaDB 10.x support

Compatibility With All Database Engines

All MySQL engines

InnoDB only

Security

4.33

5.00

Security/Encryption at rest

Can leverage existing cloud or filesystem encryption technologies

Built in

Security/Encryption in flight

Supported for all traffic

Compliance (PCI, HIPAA)

Can be used in security/privacy standards

Management and Monitoring

4.80

4.20

Deployment

Easy to deploy

Cluster Management

Easy management via web-based GUI

UI, region specific - can only view a single region at a time

Monitoring - GUI

Full GUI based cluster monitoring. Performance monitoring with Prometheus

Full monitoring, though enhanced monitoring is charged extra

Alerts / Notifications

Integration with notifications systems

Alerting via AWS api, must be configured

OS/DB Patching

Assistance with scripts, no downtime

Automatic, however will cause interruption

Real-time Data (Analytics, Data Lake, Reporting)

5.00

1.00

Replication Into Analytics

Real-time to the data analytics engines (AWS Redshift, Hadoop, Vertica, many others)

AWS Redshift only, using external tools

Filtering

Complex filtering at schema and data level

Only basic MySQL replication

Cluster Aware

Replication stream not interupted by loss of a single cluster node

External tools need to be made aware

Support Services

24/7 Support

5.00

2.00

24/7 Support

24/7 support with high, near 100% customer satisfaction

Basic support > 12 hours, business-critical Enterprise support extremely expensive

Fast Support Reponse Time

The fastest in the industry, averaging less than 3 minutes for Urgent support request.

Unless paying very high Enterprise support fee

Support Team Experience

Each member has 20+ years experience on DBA and Site Reliability

Unknown who will be providing support, often lower-level call center member with limited or no DBA experience

Intangibles

MySQL Version Support

5.00

2.00

MySQL Community

Fully supported, all versions up to 8.0

'MySQL Compatible', 5.6, 5.7, 8.0 (most) features only

MySQL Enterprise

Fully supported, all versions up to 8.0

Not supported

MariaDB

Fully supported, all versions up to 10.5

Not supported

Percona Server

Fully supported

'MySQL Compatible', 5.6 and 5.7 features only

Infrastructure and Platform

5.00

0.75

On-premises Deployment

Fully supported

Not supported

Cloud Deployment

All cloud environments (AWS, Google Cloud, MS Azure) are supported

AWS only

Multi-Cloud

Fully supported

Not supported

Hybrid-Cloud

Fully supported

Not supported

Product Maturity

5.00

3.67

Complete

One solution for HA, DR & Geo-Scale

Mainly for HA, DR only with Aurora Global, no Geo-Scale

Integrated

Comes with Manager, Proxy & Replicator

RDS Proxy is rudimentary and costs extra

Fully Tested

Proven in business-critical customer environments

Widely deployed, testing methodology unknown

Total Cost of Ownership (TCO)

4.33

3.33

Low Subscription Cost, High Value

Instance size does not matter, non-linear pricing (additional clusters are at increasingly lower cost), no cost TrueUps

Instance size based pricing, Linear pricing

24/7 Support

Included

Enterprise-level support available at a very high cost

Few DBA Resource Needed

Large number of clusters can be managed with small number of DBAs

Although, the 'managed' DBaaS still needs DBAs

Tungsten Clustering

Tungsten Clustering is the only complete, fully-integrated, fully-tested MySQL HA, DR and geo-clustering solution running on-premises and in the cloud combined with industry-best and fastest, 24/7 support for business-critical MySQL, MariaDB, & Percona Server applications.

It allows enterprises running business-critical MySQL database applications to cost-effectively achieve continuous global operations with commercial-grade high availability (HA), geographically redundant disaster recovery (DR) and geographically distributed active/active.

Continuous Operations

Local clustering for enterprise-grade high availability (HA) and multi-site clustering for best-in-class disaster recovery (DR).

High Availability

Automated local failover, automated recovery and failback for MySQL and MariaDB clusters.

Disaster Recovery

Composite “clusters of clusters” spread across two or more sites or cloud data centers for a geographically distributed managed mesh of MySQL or MariaDB clusters.

Global Scaling

Geo-distributed and global clusters that are locally and globally aware. Serve a geographically-distributed user base with high availability and performance.

Interested to Learn More? Get in Touch!

Please use the button below to contact us. We will reply within 24 hours.

Related Resources

Read the Comparison Blog

Amazon Aurora MySQL - High Noon with Tungsten Clustering for MySQL High Availability (HA), Disaster Recovery (DR) and Geographic Distribution

This is the first blog in our ‘High Noon’ comparison series in which we look at the main solutions for MySQL high availability, disaster recovery and geographic-distribution. We focus on highly available, geo-scale, multi-region MySQL for business critical applications as directly compared to Amazon Aurora. Tungsten Clustering makes it easy to deploy and manage highly available MySQL clusters with any version of MySQL, MariaDB, and Percona Server for MySQL on-premises, in the cloud, hybrid-cloud or multi-cloud.

Matthew Lang

Feb 09, 2021

High Noon Competitor Comparisons, Database-as-a-Service, MySQL High Availability, Tungsten Clustering

Watch the Comparison Presentation

MySQL HA/DR/Geo-Scale - High Noon #1: AWS Aurora

This first installment of our High Noon series of on-demand webinars is focused on AWS Aurora. It looks at some of the key characteristics of AWS Aurora and how it fares as a MySQL HA / DR / Geo-Scale solution, especially when compared to Continuent Tungsten Clustering.

Matthew Lang

Upgrade From the Following Solutions

Do It Yourself Solution
(such as with ProxySQL and Orchestrator)

Industry-Best Continuent MySQL Support

24/7 MySQL support

Service is Always Available. Experts are on-duty 24/7/365 no matter which timezone you are in.

Highly Experienced

Our support engineers have 15 or more years of MySQL DBA and Site Reliability experience.

Quick Response

24/7 support comes with 1-hour SLA, with response times for urgent requests averaging less than 3 minutes.

Highly Available

MySQL uptime measured in months or years.

Why Continuent?

Our customers are leading SaaS, e-commerce, financial services, gaming and telco companies who rely on MySQL and Continuent to cost-effectively safeguard billions of dollars annual revenue.

Learn more

Tungsten Clustering vs. Amazon Aurora

Amazon AWS Aurora for MySQL Comparison

Use Cases Supported

Features & Benefits

Support Services

Intangibles

Tungsten Clustering

Continuous Operations

High Availability

Disaster Recovery

Global Scaling

Interested to Learn More? Get in Touch!

Related Resources

Read the Comparison Blog

Amazon Aurora MySQL - High Noon with Tungsten Clustering for MySQL High Availability (HA), Disaster Recovery (DR) and Geographic Distribution

Watch the Comparison Presentation

MySQL HA/DR/Geo-Scale - High Noon #1: AWS Aurora

Upgrade From the Following Solutions

Industry-Best Continuent MySQL Support

24/7 MySQL support

Highly Experienced

Quick Response

Highly Available

Why Continuent?

These Companies Trust Our Solutions