Recently, Google Cloud SQL (RDS on Google Cloud) has made a major product adjustment and release: the original Cloud SQL is divided into two versions, namely "Enterprise" and "Enterprise Plus" versions. This article outlines the similarities and differences between the two versions to help everyone better understand Google Cloud SQL.
"Enterprise" and "Plus" editions of Cloud SQL
Overall:
1. Google Cloud SQL is equivalent to RDS of AWS or Alibaba Cloud platform, and Cloud SQL for MySQL is equivalent to AWS RDS for MySQL;
2. After this adjustment, the original Cloud SQL for MySQL/PostgreSQL is all changed to the Enterprise version, the price and the product itself have not changed, but the name has changed;
3. The newly added Enterprise Plus version provides higher product capabilities, such as the Data Cache capability provided by hardware/software optimization, which can improve the read and write performance of MySQL; Enterprise Plus also provides larger specifications, more Flexible CPU/memory ratio and other capabilities;
4. The price of Enterprise Plus resources is about 30% more expensive than the Enterprise version, and the Data Cache space is charged extra;
5. In essence, it is guessed that Enterprise Plus uses a newer and stronger model with very strong local storage capabilities (NVMe SSD or other storage devices may be used locally), and this capability is used at the database layer to enable users to obtain better Ability to read and write;
6. Judging from the pace of product releases, this year GCP (Google Cloud) has increased its investment in Cloud SQL, which should be similar to Alibaba Cloud now. In the past, in Google's data product system, Spanner, AlloyDB or BigTable were the products with the largest investment, while Cloud SQL seemed not to be taken seriously. Now it is probably found that no matter how advanced the Spanner and AlloyDB architectures are, the simplest Cloud SQL has always been the first choice for more users;
7. This name change and adjustment feels quite confusing. For Cloud SQL for MySQL, the difference between "Enterprise" and "Enterprise Plus" is quite obvious, one has Data Cache and the other does not. But for PostgreSQL, the difference between the two is more like a simple limitation on the use of resources (maximum specification, cpu memory ratio, PITR time limit, etc.). In addition, currently SQL Server does not distinguish between these two types. Therefore, the overall feeling is relatively confusing. It may be simpler to add an I/O-Optimized option to AWS Aurora. GCP's "simplification into complexity" approach may have something to do with business pressure. It is more like a product decision made for the concept, and ultimately makes users pay for the confusing product name and positioning. This kind of thing is common in large companies. It still appears frequently in enterprise-level products. If you can't make any breakthroughs, then let's make some breakthroughs in the product name (this is an irresponsible "conjecture", please refer to it with caution).
Currently, "Enterprise Plus" uses a special performance-optimized model, and the version only supports the latest MySQL 8.0 and PostgreSQL 14 and 15 versions. More specifically, this version:
- The maximum specifications can reach 128vCPU and 864GB memory
- For MySQL, it has Data Cache capability, which reduces read and write performance through local high-speed SSD devices.
- At the same time, in order to improve performance, more adaptive adjustments are made to the parameters of the instance.
product price
Let’s take a look at the price comparison between Enterprise and Enterprise Plus instances. Here is a comparison of CPU/Memory prices for a single node and Tokyo area:
As you can see, the Plus version is about 30% more expensive. In addition, Data Cache requires additional charges. in addition:
If it is a dual-node high-availability instance, the price needs to be doubled.
The capacity of the data cache of the instance is fixed and matched with the instance specification, and the user cannot configure it. For example, the data cache size corresponding to the 8vCPU/64GB instance is 375GB
More about Data Cache
Data Cache not only provides local SSD capabilities, but also expands the database's Buffer Pool. If Data Cache is enabled, read requests will be read from memory first, then Data Cache, and finally instance storage. When data is written, a copy is also written to the Data Cache.
How to open Data Cache
Data Cache is optional and needs to be turned on when the instance is created. The opening option is slightly hidden, specifically: when creating an instance, it can be opened in the Machine Configuration under Customize your instance, as shown in the figure below:
at last
The three major overseas cloud vendors (AWS, Azure, GCP) all have some unique features. AWS was the first to do this, and its relatively standard product names can be regarded as industry de facto standards, but neither Azure nor GCP "respect" this industry leader in product naming. For example, Azure's RDS is called "Azure Database For x", and Google Cloud is called "Azure Database For x". Cloud SQL for x".
Now each company is also increasing investment in the database, as time goes by, so that the product form of each company is becoming more and more different, which is relatively confusing for users. And if users want to switch between multiple clouds, the cost is also increasing. This is probably one of the potential goals of various cloud vendors.
Author: orczhou@ninedata is a database product expert from Jiuzhang Arithmetic. Nine Data NineData is a new generation of cloud-native intelligent data management platform, including data replication, SQL development, data backup and data comparison and other functions. NineData adopts leading cloud native and AIGC technologies to provide an intelligent data management platform for architecture design in the cloud and AI era. As the industry's first platform that supports seamless integration of cross-cloud and local IDC, it can help customers easily complete data migration to the cloud, real-time transmission of cross-cloud data, ETL, data backup, enterprise-level intelligent SQL development, database development specifications, production changes, and sensitive Data management and other functions to make customer data use more secure and efficient. NineData - Let everyone make good use of data and cloud - Nine Chapter Arithmetic