Cloud, in general, has matured a lot during the last decade, providing a fresh wave of confidence to organizations still relying on legacy systems for their data and analytics requirements. There are a plethora of options for an organization to adopt from, depending on their immediate or specific needs. This article addresses anyone or any organization looking into data warehousing options which are available in the cloud and highlight Snowflake Cloud Data Platform, and how it can be a great choice while thinking about migrating to a new cloud data warehouse.
The cloud data warehouse market is a very competitive space, but is also defined by the unique offerings of a provider. AWS Redshift, Azure, SQL data warehouse, Google BigQuery are great options in a rapidly evolving data warehousing market, which is valued at over 18+ billion USD. However, it seems, Snowflake has changed the game.
Snowflake is a cloud data platform in its true sense, it didn’t evolve from an on-premise solution or using an existing database engine. Snowflake data warehouse works on a SaaS, fully managed and pay-as-you-go model making adoption quick and easy. Most of all Snowflake architecture is what makes it so special.
Snowflake data warehouse has a hybrid of shared-disk and shared-nothing database architectures. Benefitting from both architecture, Snowflake uses a central repository for persisted data which is accessible from all compute nodes and processes query using MPP compute clusters.
Snowflake’s Database Storage Layer is synchronously replicated across multiple disk devices and three separate availability zones – in the same region – ensuring high data availability.
Snowflake’s Virtual Warehouse Processing layer consists of one or many virtual warehouses completely depending on an organization’s requirements and configurations. These virtual warehouses are independent compute clusters that do not share compute resources, hence providing an easy option to separate certain workloads to avoid contention and allow infinite horizontal scalability. It is recommended to isolate compute or virtual warehouses by Activity (data load, reporting, exploration, etc.) or Data Source (SFDC, SAP, Social etc) or Business Units (Sales, Marketing, Finance etc.).
Cloud Services Layer is a collection of global services that coordinate activities across the data warehouse. These services are fully maintained and managed by Snowflake for metadata, security, query compilation, optimization and warehouse management.
Snowflake emerged as a revolutionary player in the market completely isolating the storage from the compute, providing auto and full elasticity as a managed service with pay-as-you-go thus becoming a market leader.
Top 10 reasons for migrating to Snowflake:
- Fully managed SaaS cloud data warehouse with a pay-as-you-go pricing model (More details here)
- Isolated storage & computes with infinite scaling (vertical for performance and horizontal for concurrency) (More details here)
- ANSI SQL compliant with complete support for analytical functions with automated query optimization. (More details here)
- Supports structured and semi-structured (Parquet, CSV, JSON, AVRO, XML, ORC, etc.) (More details here)
- Cross-Cloud Capabilities with inbuilt data compression & encryption. (More details here)
- Hierarchical Key Model Encryption to ensure Data Security & Compliance for HIPAA, GDPR, and PCI-DSS (More details here)
- Auth using SSO (AD, OKTA, SAML) with an MFA (Duo) (More details here)
- Advanced features like data sharing and zero-copy clone preventing data export (I/O) and data replication for multi-vendor engagement or downstream application integration improving data access traceability (More details here)
- Great technology support (hands-on technical tales field engineers) & active community
- Amazing partner connects (ELT/ETL, Data Fabric/Virtualization, Reporting/Visualization) using various drivers (JDBC, ODBC, Go, Node.js) and connectors (Kafka, Spark, Python) (More details here)
On top of these benefits, Snowflake is rapidly evolving into a cloud data platform. In 2019, Snowflake introduced a Data Exchange platform which in effect is a marketplace for thousands of datasets updated regularly. Without any ETL, data movements or storage costs consumers can access thousands of datasets to derive better business insights. Snowflake masterly positions itself with an extremely powerful ecosystem, providing a variety of integration touch-points across the modern technological landscape enterprise. (More details here)
Saama Analytics has been a huge consumer and contributor to Big Data Landscape. Partnered with Snowflake in early 2019, we have successfully migrated several legacy and cloud-based systems to Snowflake including SQL Server, Teradata, Oracle, HDFS, Hive and AWS Redshift.
Realizing the potential of Snowflake, at Saama Analytics we have developed Snowflake Accelerators that help drastically improve time to value. If you are looking at modernizing your enterprise adopting the latest tools and technologies, happy to connect.