The latest Oracle Exadata X8 is the 9th generation. Every generation has been laser focused at raising the expectations bar for increased performance and automation; reducing database administrator (DBA) expertise, skills, knowledge, and experience requirements; radically reducing wasted time for both the DBA and user; and continually lowering the total cost of ownership (TCO). Each generation has wildly succeeded in raising that bar increasing the distance between Exadata and every other database system or white box. Exadata X8 has raised the bar so high that it’s out of sight for the nearest competitor. A closer look shows why.
For decades, databases have been an essential mission critical application running on servers. DBAs trained for years to perfect database performance tuning, database optimization, troubleshooting, healing, prioritization, and more to make critical applications running on the database run faster and more reliably. Each generation of server, storage, and networking would increase database performance. The past decade has seen exponential database growth while storage performance has seen a massive increase with the proliferation of low latency flash SSDs, NVMe SSDs, and storage class memory (SCM) drives.
Concurrently there have been hiccups in Moore’s law. It has slowed to a crawl. Apparently, there are limits to how small the lithography can go. CPU performance increases have been marginal focusing more on adding cores instead of performance. This is true for both all CPUs – x86, RISC, even ARM. This has resulted in the compute server becoming the database performance bottleneck. The compute server vendors have attempted to solve this problem with hypervisors, clustering, and sharding. Each having some but limited success in scaling performance and capacity while introducing other problems. Problems such as increased complexity, vendor management, patching, complicated lengthy troubleshooting and problem resolution. All of which causing user frustration.
Another problem has been the massive loss of DBA knowledge, skill, and experience as the baby boomer generation retires in large numbers. College graduates lack the underlying expertise and have long DBA learning curves. Databases must be simpler to use, operate, tune, manage, patch, troubleshoot, etc., to be utilized efficiently and effectively. Meanwhile, the number of distinct database types (relational, key value, graphical, time series, object, columnar), open source, cloud database services, and specialized databases have exploded into IT organizations ecosystems. This plethora of database products and services demands expanded DBA programming skills in SQL, JSON, XML, r, and q command languages; as well as DBA knowledge and expertise in as many processes as there are database types; plus, extensive costly training. Multiple databases introduce other challenges and problems as well. Analyzing raw data in different types of databases too often require the data be moved between the databases. That requires a manually labor-intensive extract transfer and load (ETL) requiring a lot of time. It’s difficult and takes so much time that the information derived by the analytics is likely out-of-date by the time the data is moved and analyzed. And it’s not a one-time thing. There’s a reason Gartner has said 85% or more of big data projects fail. One of the most onerous is the requirement to move data between databases.
The most common workaround to the multi-database problem is to create islands of data for each database type. This causes a nontrivial amount of data duplication that cannot be solved at the storage level. That explosive data expansion complicates data center infrastructure, storage, servers, data protection, disaster recovery, business continuity, and cost. Keeping all the data synchronized and up to date is also problematic at best and costly in both human and IT assets.
All of these problems combined are causing a brisk unsustainable rise in database costs. IT organizations are demanding a faster, less complicated, more automated, lower cost answer.
Many IT pros are surprised that Oracle, the world-wide leader in databases for decades, solves all of these problems and more with the Exadata X8. They shouldn’t be. Oracle has been doing this with Exadata releases for over a decade. Exadata X8 raises standards to a level previously not thought possible.
It solves the compute server bottleneck problem by making the Oracle Database hardware aware and the server, storage, and network Oracle Database aware. Then offloads many of the database processes including SQL, XML, JSON, encrypt, decrypt, RMAN backup filtering, fast file creation, many in-database analytics and machine learning (ML) functions to the storage processors. Those processors are much closer to the data shortening speed of light latencies speeding up results. That offload additionally frees up compute resources to do more queries, higher-level AI/ML, and more analytics. It also makes clever use of flash caching to put more database processes in-memory, making scans much faster, automatically reducing I/O, and automates I/O prioritization. Internal use of RDMA makes internodal RAC performance as fast as if it were in a single node while eliminating the ills of a distributed or clustered database.
The implementation of built-in algorithms and AI/ML has automated database operations to a level not seen before. Operations such as Automatic Indexing based on policies, machine learning, and reinforced machine learning. Indexing that takes a skilled experienced DBA hours or days is accomplished in seconds to minutes. And the automated indexes are more efficient and significantly faster than the ones created by the most experience DBAs. There are more than 60 unique features in Exadata X8 that are simply not available in any other platform. All of these features are designed to increase performance, automation, simplicity, reliability, availability, and database effectiveness.
Exadata X8 also solves the scalability, multiple database, and multiple database type conundrums. Based on the latest and greatest all-inclusive Oracle Database, it supports up to 4,000 pluggable databases (PDB) in a multitenant container database environment. It also supports databases in the hundreds of TBs to PBs of data. It effectively supports 10-15x that amount of data because of the Hybrid Columnar Compression (HCC). Each Exadata scales from 1/8th rack to 18 full racks. It can scale further with external switches supporting RDMA.
Exadata X8 solves the multiple database type problem by incorporating those database types into the Oracle Database. It supports relational, document, key value, graphical, time series, object and more. This enables each database type to access the same data without an ETL or data copy.
But what about the cost issue? Once again Oracle has attacked this problem head-on. Oracle’s co-engineering of the Exadata X8 platform with the Oracle database measurably reduces the amount of hardware and IT infrastructure required to run optimized. Exadata X8 is backwards compatible with multiple generations of Exadatas and will therefore be compatible with multiple generations of future Exadatas making growth and tech refresh a non-event. Oracle also offers Exadata X8 in three ways:
Cost comparisons with do it yourself (DIY) white box and named vendor implementations show a much lower TCO for Exadata X8. The medium average Exadata advantage was much more than 50%. When Exadata’s performance advantage is considered, the difference is multiple orders of magnitude in favor of Exadata.
Exadata X8 solves today’s extensive database problems, is faster, far more automated, more complete, and much more cost effective than any other database platform. This is why Oracle Exadata X8 makes DIY Oracle Database hardware systems obsolete and why Oracle remains number 1 in databases.
Oracle’s Exadata X8 convergence is at the PhD level. Whereas everyone else is in elementary school, including Dell-EMC, Nutanix, and HPE.
About Dragon Slayer Consulting: Marc Staimer, as President and CDS of the 21-year-old Dragon Slayer Consulting in Beaverton, OR, is well known for his in-depth and keen understanding of user problems, especially with storage, networking, applications, cloud services, data protection, and virtualization. Marc has published thousands of technology articles and tips from the user perspective for internationally renowned online trades including many of TechTarget’s websites and Network Computing and GigaOM. Marc has additionally delivered hundreds of white papers, webinars, and seminars to many well-known industry giants such as: Brocade, Cisco, DELL, EMC, Emulex (Avago), HDS, HPE, LSI (Avago), Mellanox, NEC, NetApp, Oracle, QLogic, SanDisk, and Western Digital. He has additionally provided similar services to smaller, less well-known vendors/startups including: Asigra, Cloudtenna, Clustrix, Condusiv, DH2i, Diablo, FalconStor, Gridstore, ioFABRIC, Nexenta, Neuxpower, NetEx, NoviFlow, Pavilion Data, Permabit, Qumulo, SBDS, StorONE, Tegile, and many more. His speaking engagements are always well attended, often standing room only, because of the pragmatic, immediately useful information provided. Marc can be reached at firstname.lastname@example.org, (503)-312-2167, in Beaverton OR, 97007.
Go to: Oracle Exadata