Credited: snowflake.com
1 0
Read Time:2 Minute, 44 Second

Data protection and High Availability (resiliency) are the core capabilities of Snowflake cloud-built data warehouse-as-a-service. Snowflake assures resiliency and data consistency in the event of node failures, as it automatically distributed across three availability zones. Each data center in AZ contains a huge pool of low-cost, commodity hardware . Moreover, there is also no impact on write (or read) performance in case of node failures. As you ingest your data, it is synchronously and transparently replicated across availability zones. This protection is automatically extended from Snowflake to customers, at no added charge.

Snowflake Time Travel enables you to recover data from any point, up to 90 days.  In addition, it’s all accomplished automatically. 

Few Terms to be understand:

Region: Region is a separate geographic area like us-east,us-west,Europe,Asia Pacific etc…

Availability Zones(AZ) : Availability Zones are multiple, isolated locations within each Region. However, They provide inexpensive, low-latency network connectivity to other Availability Zones in the same AWS Region. If you distribute your instances across multiple Availability Zones and one instance fails,  an instance in another Availability Zone can handle requests. This is like an emergency load balancer without using an actual load balancer.

AWS AZ

An Availability Zone (AZ) contains one or more Data Centres (DC’s) DC’s are located close to each other so that connectivity is quick and latency is as low as possible.

Snow URL

When you request a Snowflake account, either through self-service or a Snowflake representative, you can choose the region where the account is located. For example, you can decide to locate an account in a particular region on a particular cloud platform to address latency concerns. If latency is a concern, you should choose the available region with the closest geographic proximity to your end users.

Virtual warehouses and query resiliency:

Virtual warehouses and query resiliency:

Query execution tasks are processed in the virtual warehouses. If a single compute instance fails within a virtual warehouse, Snowflake automatically replaces that instance and completes the query without disruption. Moreover, Virtual warehouses do not have permanent storage of data. Instead, they cache data to improve query performance. Loss of any number of nodes in a virtual warehouse, including the loss of an entire virtual warehouse, does not cause a data loss since the persistent copy of data resides in the cloud object stores.

Virtual warehouses and AZ:

However, For performance reasons, each virtual warehouse runs within a single availability zone. If Snowflake detects that an availability zone has become unavailable, the cloud services layer can re provision the impacted virtual warehouse in a different zone and restart queries that have been in process to ensure that they complete successfully.

Moreover, if you can connect to your Snowflake account, you can query the following context functions to identify the region and account locator for the Snowflake account you are connected to:

SELECT CURRENT_REGION;

SELECT CURRENT_ACCOUNT;

References: https://www.snowflake.com/blog/how-to-make-data-protection-and-high-availability-for-analytics-fast-and-easy/

Image Credited: snowflake.com

Average Rating

5 Star
0%
4 Star
0%
3 Star
0%
2 Star
0%
1 Star
0%

Leave a Reply

Your email address will not be published. Required fields are marked *