Oracle’s new Autonomous Data Warehouse Cloud is democratizing the world of Data Warehousing so that literally anyone can gain access to industrial strength data analytics.
Gone are the days when millions of dollars and months of time are needed to setup a Data Warehouse using Teradata, Exadata, or some other system just to gain insight into your business data. Users can now easily and quickly gain access to the fastest, most secure, and most capable analytic platform in the Cloud.
Autonomy = Eliminating DBA Tasks
Oracle’s Autonomous Data Warehouse Cloud eliminates back-end DBA tasks such as software installation, patching, upgrades, system optimization, system tuning, database backup, and other mundane but time consuming tasks that don’t drive business value. Oracle has taken control of the back-end system and turned it over to Artificial Intelligence algorithms, allowing database Administrators to focus on more high value work such as data modeling, data architecture, and optimizing business processes.
Database Administrators should NOT worry about Autonomous Data Warehouse making their job obsolete. Oracle has been automating administrative tasks for decades, making the role of DBA’s even more valuable. DBA’s will be able to put even greater focus on business processes, data modeling, data architecture, and performance of business processes rather than mundane tasks like patching software.
One Warehouse, Many Users
In typical Oracle parlance, one Data Warehouse can serve many users who often share data, sections of data, or what is known as subject areas within the same Warehouse. Sales, Marketing, Customer Service, and other users can have their own separate data within the Warehouse, but can also easily share data between them.
Fast & Flexible Data Loading
Oracle Autonomous Data Warehouse Cloud supports all of the leading data loading or ETL products that customers are familiar with. The following 3 sources can be used for loading data directly into your Data Warehouse:
- Load from on-premise sources
- Load from Amazon Cloud Storage
- Load from Oracle Object Cloud Storage
Autonomous Data Warehouse also allows data to be loaded using simple SQL statements with data in these sources defined using what are known as external tables. For extremely large volumes of data, Oracle also provides data transport services.
No matter which method you choose for loading data, Autonomous Data Warehouse makes it fast, easy, and secure.
Directly Access External Data (Amazon S3, etc.)
Autonomous Data Warehouse provides the fastest access to data that has already been loaded into tables within the Warehouse, but users can also directly query data in Amazon S3, in Oracle Object Cloud Storage, as well as from on-premise data sources. These external data sources are mapped as external tables, giving users direct access to that data using simple SQL statements.
No Indexes to Create
Autonomous Data Warehouse Cloud greatly simplifies administration by eliminating the need for users to create indexes. ADWC does have indexes, but users don’t need to create them. ADWC provides excellent performance even for single row queries.
Internal Performance Technologies
Automated Data Warehouse delivers unsurpassed performance through several technologies including the following:
- Automatic Multi-Level Caches
- Active Storage
- Automatic Data Indexing
- Automatic Optimizer Statistics
Users need not be concerned with these technologies beyond knowing they exist and give excellent performance. Prior to the advent of the Autonomous Data Warehouse Cloud, database administrators would often spend countless hours using manual methods to tune what is now done autonomously.
Automatic Data Caches
One mechanism used by ADWC is a multi-level system of data caches. As data is used more and more often, ADWC stores or “caches” that data in faster layers of caches. Data automatically moves from spinning disk to Flash, to memory in the Active Storage layer, to DRAM caches in the CPU processing layer. Data is automatically brought into In-Memory data stores, and results are even cached to speed later access to the same data. The size and uses of each cache is handled autonomously by ADWC.
Data in the Autonomous Data Warehouse is housed in storage, of course, but not in convention disk storage as with other Cloud services. Instead, ADWC uses an “active storage” layer, which transparently brings additional processing power to handle user queries without any user effort.
Active Storage in the Autonomous Data Warehouse simply makes warehouse queries run faster without any user intervention or performance tuning. Exadata users know this as SQL Offloading, but it simply means that queries run faster by sending query work down to the storage layer for processing. This is a time-tested approach in computer science, shipping processing logic to the data, rather than shipping data to the processing logic. Whenever processing happens closer to the data (without moving the data), users simply see faster processing.
Automatic Data Indexing
Anyone who knows even the basics of databases and data warehousing knows that indexes make data access faster. People often wonder if Autonomous Data Warehouse doesn’t allow users to create indexes, how can it possibly be fast? The simple answer is that ADWC does have indexes, but they are created internally and automatically. For example, ADWC uses a feature pioneered in Exadata called “Storage Indexes” that automatically creates indexes in the Active Storage layer on ADWC.
Automatic Optimizer Statistics
Users of conventional Oracle Data Warehouses on Exadata and other platforms might have heard about optimizer statistics from their DBA. Optimizer statistics can be difficult to manage and often have a critical impact on performance of business processes. The Autonomous Data Warehouse automates the collection and management of optimizer statistics to ensure optimal performance.
Pay as You Go – Compute vs. Storage
Customers pay for storage space on the Autonomous Data Warehouse Cloud Service in increments of 1TB. Compute power is charged on an hourly basis by the OCPU, and can be easily scaled up or down as needed using. Customers can completely TURN OFF the compute power if it’s not being used (such as during off-hours, on a weekend, etc.).
Encrypted by Default
Autonomous Data Warehouses are encrypted by default, providing protection for user data regardless of source or sensitivity. Oracle has always been a leader in data security since its inception, and those security technologies are built into the Autonomous Data Warehouse Cloud.