Snowflake Summit 2022
Kazuhiko Fujimoto (Editor)
2022-06-15 10:10
Writing in notenote
What will be changed by automation of configuration management? Optimize cumbersome IT operations.
Is it really that easy? What are the key points for successful low-code development?
Platforms for Full-Scale Kubernetes Utilization
Many white papers including case studies, product information, surveys and reports
Snowflake is holding its annual conference "Snowflake Summit 2022" in Las Vegas, NV from June 13-16, 2010 (US time). In conjunction with the conference, a series of new features for the company's data cloud product were announced.
Native Python Support and Expanded Data Access
First, the company announced enhancements to its developer features, centered on stronger Python support. Specifically, a public preview version of Snowpark for Python is now available, and native integration with Streamlit, which is currently under development, will enable a faster application development cycle. In addition, the company will enhance its new streaming data integration and make open-format data and data stored on-premise available in the data cloud.
-PR-.
Snowpark is a developer framework for building scalable pipelines, applications, and machine learning (ML) workflows directly on Snowflake using any development language or library. Snowpark for Python runs on Snowflake's computing infrastructure, as do Snowflake pipelines and applications.
Image 1
In addition, the following new features will complement Snowpark for Python
Snowflake Worksheet for Python (private preview): directly on Snowflake's "Snowsight" user interface, using Python and Snowpark's DataFrame API for Python, pipelining, machine learning (ML) Develop models and applications, and streamline your development efforts with code auto-completion and the ability to generate custom logic in seconds
Snowflake integration with Streamlit (under development): Python-based application development capabilities will be built directly into Snowflake. Users will be able to develop interactive applications, securely share data, engage in fast, iterative development cycles, and collaborate with business teams to increase development impact
Large memory warehousing (under development): users can securely perform memory-intensive tasks such as feature engineering and model training with large data sets using open source Python libraries made available through integration with Anaconda Enables users to safely perform memory-intensive tasks such as feature engineering and model training on large data sets
SQL machine learning, including time series prediction (private preview): SQL users can incorporate ML-based prediction into their daily business intelligence (BI) and analytics to improve the quality and speed of decision making.
In addition, Snowpipe Streaming (Private Preview) for serverless streaming data ingestion and Materialized Table (under development) to simplify declarative streaming data transformation were announced. The open source table format "Apache Iceberg" is now available for external storage, and on-premise storage systems such as Dell Technologies and Pure Storage supported by Snowflake Data stored on on-premise storage systems supported by Snowflake, such as Dell Technologies and Pure Storage, can now also be accessed.
Kazuhiko Fujimoto (Editor) 2022-06-15 10:10
Writing in notenote
Optimize IT infrastructure operation and management! Why is "configuration management" so important?
Is it really that easy? What is the key to successful low-code development?
Kubernetes is not enough. How to compensate for missing functions?
Many white papers including case studies, product information, surveys and reports
Data Cloud Enables App Development, Monetization, and Deployment
Next, the company announced a new platform for developers to develop, monetize, and deploy data-intensive applications on the data cloud.
The Native Application Framework, currently in private preview, allows developers to leverage Snowflake features such as stored procedures, user-defined functions (UDFs), and user-defined table functions (UDTFs) to develop applications. The company is also working on integrating its functionality with Streamlit, an application framework for Python.
-PR-.
Developers can monetize the applications they build in the "Snowflake Data Marketplace"; as of April 30, 2022, there are 6300 companies using Snowflake. The native application framework builds on the data cloud's availability, resiliency, and security posture, allowing developers to focus solely on developing functionality without worrying about work and operational burdens. The sold application runs within the customer's Snowflake account. There is no need to move or share data, and developers do not have to manage sensitive customer data.
Customers and partners such as LiveRamp and Informatica are using the native application framework to develop applications for a variety of use cases, including cloud cost management, identity data verification and matching, and data collection.
Image 2
Unistore" for Transaction Processing
Snowflake announced the launch of Unistore, a new workload that combines transactional and analytical data on a single platform.
Unistore is a new workload for combining transactional and analytical data on a single platform. Previously, transactional and analytical data were siloed, complicating the movement of data between systems and compromising the speed needed for modern development; Unistore enables data clouds to be leveraged for transactional workloads.
Snowflake introduced "Hybrid Table" as a new feature to support Unistore. The ability to manipulate single rows (records) at high speed will enable the development of transactional applications on Snowflake. They can also quickly perform analysis of transactional data and combine existing "Snowflake Tables" for a consolidated view of data.
Christian Kleinerman, Snowflake's senior vice president of product, said, "Unistore is the foundation for new innovations in the Snowflake Data Cloud. Just as we have been shattering customer stereotypes about data lakes and data warehouses, Unistore will usher in the development and deployment of next-generation applications on the data cloud.
Image 3
New Cybersecurity Workload Offerings
Snowflake also announced the launch of a new cybersecurity workload offering to help corporate cybersecurity teams better protect themselves with data clouds.
According to the company, current security architectures built around older security information management systems (SIEMs) are not designed to handle the volume and types of data needed to stay one step ahead of cyber threats. Due to various limitations such as data acquisition costs, retention period limitations, and proprietary query languages, the security team struggles to ensure the visibility needed to protect their company.
The company's cybersecurity workload natively processes structured, semi-structured, and unstructured logs, leveraging the processing power and elasticity of its data cloud infrastructure. Currently in private preview. The company claims that users can efficiently store large amounts of data spanning years, use the computing resources of the data cloud to perform searches, and gain insights using general-purpose languages such as SQL and Python.
By centralizing security and enterprise data in a single source of truth, companies can also use contextual data from human resources (HR) systems and IT asset management for detection and investigation to increase the fidelity of alerts, or to perform fast queries on large volumes of data. data, and execute queries at high speed against large volumes of data.
Security teams can gain comprehensive visibility into their security posture and eliminate data silos without incurring huge data acquisition and retention costs. In addition to threat detection and response, cybersecurity workloads address a wide range of use cases, including security compliance, cloud security, identity/access, and vulnerability management.
0 コメント:
コメントを投稿