Snowpark Migration Accelerator 文档¶

传统数据平台和大数据解决方案通常无法实现其主要目标：允许用户在不受大小、速度或适应性限制的情况下处理数据。Snowflake 的 Data Cloud 提供完整的解决方案，惠及所有从事数据处理工作的人员，包括：

• 数据分析师

• 数据科学家

• 数据工程师

• 商业专业人士

• 技术专业人士

迁移到基于云的领先数据平台 Snowflake 可以带来显著的优势。

• Multi-Cluster and Shared Data: Processes large amounts of data quickly and efficiently by using multiple clusters that can access the same data.

• Micro-Partitioning: Stores customer data in small, manageable chunks for better security and efficiency.

• Delivered as a Service: No need to manage or maintain the platform yourself - everything is handled for you.

• Data Platform Built for Any Cloud: Multiple teams can work with the same data simultaneously because storage and computing resources are separate.

• Improved Performance and Throughput: Works faster than traditional data processing methods.

• Support for All Data: Can work with both regular structured data and semi-structured data like JSON, Avro, XML, and Parquet files.

• Build in Your Language of Choice: Write code in Python, Scala, or Java using Snowpark's secure libraries and runtime environments.

Snowflake's Snowpark Migration Accelerator (SMA) is a user-friendly tool that helps you modernize your traditional data platform by moving it to the Snowflake Data Cloud. SMA scans your Python and Scala source code that contains Spark API calls, analyzes it, and creates an inventory. It then converts this Spark code into equivalent code that uses the Snowpark API. After the conversion is complete, you can start taking advantage of Snowflake's features.

Please note: The Snowpark Migration Accelerator (SMA) is not an official Snowflake Inc. product and is not included in the Snowflake Service. SMA is provided as-is under its own terms. Snowflake's support team does not provide assistance for SMA, and it is not covered by Snowflake's standard support and service level agreements. For more information, please contact sma-info@snowflake.com.