Engineered For Fast and Reliable Deployment: DATA SHEET / Oracle Exadata Database Machine X8M-2
Engineered For Fast and Reliable Deployment: DATA SHEET / Oracle Exadata Database Machine X8M-2
Engineered For Fast and Reliable Deployment: DATA SHEET / Oracle Exadata Database Machine X8M-2
A brand new high-bandwidth low-latency 100 Gb/sec RDMA over Converged Ethernet (RoCE)
Network Fabric connects all the components inside an Exadata Database Machine. Specialized
database networking protocols deliver much lower latency and higher bandwidth than is possible with
generic communication protocols for faster response time for OLTP operations and higher
throughput for analytic workloads. External connectivity to the Exadata Database Machine is via Key Benefits
standard 10 Gb/sec or 25 Gb/sec Ethernet. • Pre-configured, pre-tested
system optimized for all
Exadata Database Machine is the most versatile database platform. The Exadata X8M-2 database applications
Database Machine uses powerful database servers, each with two 24-core x86 processors and • Uncompressed I/O bandwidth of
384 GB of memory (expandable up to 1.5 TB). Exadata also uses scale-out, intelligent storage servers up to 560 GB/sec per full rack
available in two configurations – High Capacity (HC) or Extreme Flash (EF). HC Storage Servers have from SQL
• Ability to perform up to 16M 8K
four NVMe PCI Flash cards each with 6.4 TB (raw) Exadata Smart Flash Cache and twelve 14 TB
database read I/O operations, or
7,200 RPM disks. EF Storage Servers have an all-flash configuration with eight NVMe PCI Flash 5.1M 8K Flash write I/O
drives, each with 6.4 TB (raw) storage capacity. Exadata X8M HC and EF Storage now include operations per second in a
persistent memory, further boosting capacity and performance. HC and EF Servers receive twelve 128 single rack
GB Intel® Optane™ DC Persistent Memory modules as a new tier between DRAM and flash. Exadata • Easily add compute or storage
combines persistent memory with innovative RDMA algorithms that bypass the network and I/O servers to meet the needs of
stack, eliminating expensive CPU interrupts and context switches, reducing latency by 10x, any size application
• Scale by connecting multiple
from 200µs to less than 19µs.
Exadata Database Machine
X8M-2 racks or Exadata X8M
The minimal configuration of an Exadata Database Machine consists of two database servers and
Storage Expansion Racks. Up to
three storage servers, which can be expanded into elastic configurations adding more database 18 racks can be connected by
and/or storage servers within the same rack. Elastic configurations provide a flexible and efficient simply adding RoCE cables and
mechanism to meet any size business need. internal switches. Larger
configurations can be built with
In addition to expanding within a rack, multiple Exadata X8M racks can be connected using the external RoCE switches
integrated RoCE network fabric to form even larger configurations. For example, a system
composed of four racks is simply four times as powerful as a single rack: it provides four times the I/O
throughput, four times the storage capacity, and four times the processing power. It can be configured
as a single system or logically partitioned for multiple databases. Scaling out is easy, as Oracle Real
Related Products
Another smart new Exadata System Software feature boosts log write performance. Log write latency
is critical for OLTP performance, a faster log write means faster commit times. Inversely, any “Oracle Exadata enabled
slowdown of log writes can cause the database to stall. Unique to Exadata X8M, Exadata Smart seamless and rapid migration
PMEM Log automatically enables the database to issue a one-way RDMA log write to persistent of our cloud-based, integrated
memory. RDMA and persistent memory technologies allow the log write to occur without business applications, required
acknowledgement, and smart software places the write across multiple servers for resilience. This no major modifications, and
leads to an 8x performance increase in log writes. helped us reduce costs. With
Oracle, we dramatically
Security and management of this new tier are also automated. Persistent memory is configured improved our batch processing
speed by 241x and overall
automatically at installation time, with no user interaction required. Hardware monitoring is configured
system performance by 3x,
out of the box. Persistent memory is only accessible to databases using database access controls, enabling us to improve
ensuring end to end security of data. Deploying persistent memory in Exadata X8M is so simple, it’s operating efficiency for our
transparent. daily, critical tasks and to
enhance customer service.”
Kyoji Kato
Executive Officer and GM
Daiwa House
Exadata X8M uses a combination of scale-out storage, RDMA over Converged Ethernet networking,
database offload, persistent memory accelerator and PCI Flash to deliver extremely high performance
from memory and flash. A single rack configuration of Exadata Database Machine X8M-2, with 11
database servers and 11 Extreme Flash storage servers, using the new persistent memory
accelerator, can achieve up to 16 Million random 8K database read and 5.17 Million random 8K
flash write I/O operations per second (IOPS), which is an industry record for database workloads.
These are real-world end-to-end performance figures measured running SQL workloads with standard
8K database I/O sizes inside a single rack Exadata system, unlike storage vendor performance figures
based on small I/O sizes and low-level I/O tools and are therefore many times higher than can be
achieved from realistic SQL workloads. Exadata’s performance on real database workloads is orders
of magnitude faster than traditional storage array architectures, and is also much faster than current
all-flash storage arrays, whose architecture bottlenecks flash throughput.
Milt Simonds
HIGH CAPACITY STORAGE SERVER: TIERED DISK FLASH AND PERSISTENT Director, Enterprise Platform
MEMORY DELIVER COST OF DISK WITH SHARED MEMORY PERFORMANCE Delivery
AmerisourceBergen Corporation
The second Exadata storage option is the High Capacity (HC) Storage Server. This server includes
twelve 14 TB SAS disk drives with 168 TB total raw disk capacity. It also has four Flash Accelerator
F640v2 NVMe PCIe cards with a total raw capacity of 25.6 TB of flash memory. Exadata X8M adds
the shared persistent memory acceleration tier, twelve 128 GB Intel® Optane™ DC Persistent
Memory modules in front of flash to boost performance even more. Deployed using smart software,
Exadata Smart PMEM Cache, only the hottest database blocks are automatically cached in this new
tier. Accessible over RDMA direct from the database delivers the highest I/O rates at an extremely low
latency.
Flash in the HC Storage Server can be used directly as flash disks, but is almost always configured as
a flash cache (Exadata Smart Flash Cache) in front of disk storage behind the PMEM Cache to
deliver the best performance. Exadata Smart Flash Cache is used in-sync with PMEM Cache to
automatically cache frequently accessed data while keeping infrequently accessed data on disk,
delivering the high I/O rates and fast response times of flash with the large capacity and low cost of
disk. Exadata uniquely understands database workloads and knows when to avoid caching data
that will negatively affect overall performance. For example, if large write I/Os caused by backups
or large table scans are likely to disrupt higher priority OLTP or scan operations, those large I/Os will
bypass the flash cache and go straight to disk. Otherwise, Exadata System Software will utilize
additional spare flash capacity and I/O bandwidth to optimize performance by caching these I/Os. In
addition to automatic caching, administrators can optionally provide SQL directives to ensure that
specific tables, indexes, or partitions are preferentially retained in the flash cache.
It is common for hit rates in the Exadata Smart Flash Cache to be over 95%, or even 99% in real-world
database workloads, yielding an effective flash capacity many times larger than the physical flash. For
The automatic data tiering between RAM, persistent memory, flash and disk in Exadata provides Vaibhav Samant
Senior Vice President, IT
tremendous advantages over other flash-based solutions. Many storage vendors have developed
HDFC Bank Ltd.
flash-only arrays to achieve higher performance than traditional arrays. These flash-only arrays deliver
better performance but cannot match the cost advantages of Exadata’s smart tiering of data between
disk and flash, as the overall size of data that can benefit from flash is limited to the size of expensive
flash. And these flash arrays are unable to benefit from Exadata’s unique database-aware storage
optimization technologies. Generic data deduplication provided by some flash arrays is effective for
Virtual Desktop Infrastructure environments, but not for databases.
Exadata not only delivers much more capacity than generic all-flash arrays, it also delivers better
performance. Flash-only storage arrays cannot match the throughput of Exadata's integrated and
optimized architecture with full 100 Gb/sec RDMA over converged ethernet based scale-out network,
fast PCI Flash, offload of data intensive operations to storage, and algorithms throughout that are
specifically optimized for databases.
This storage option extends the operational and management benefits of Exadata to rarely accessed
data that must be kept online. Exadata’s Extended (XT) Storage Server is:
• Efficient – The XT server offers the same high capacity as the HC Storage server, including
Hybrid Columnar Compression
• Simple – The XT server adds capacity to Exadata while remaining transparent to
applications, transparent to SQL, and retains the same operational model
• Secure – The XT server enables customers to extend to low-use data the same Exadata
security model and encryption used for online data
• Fast and Scalable – Unlike other low-use data storage solutions, the XT server is integrated
to the Exadata fabric, for fast access and easy scale-out
• Compatible – The XT server is just another flavor of Exadata Storage server – you can just
add XT servers to any Exadata rack
With Exadata Extended (XT) Storage Server, enterprises can meet their long-term data retention
compliance requirements with the same trusted and continually validated Exadata solution, avoiding
the operational risks and costs of managing information lifecycle across multiple platforms.
One of the many unique features of Exadata System software is Smart Scan technology, which
offloads data intensive SQL operations from the database servers directly into the storage
servers. By pushing SQL processing to the storage servers, data filtering and processing occur
immediately and in parallel across all storage servers, as data is read from disk and flash. Only the
rows and columns that are directly relevant to a query are sent to the database servers.
For example, if a query is executed to identify the customers who placed sales orders over $1000 in
the month of March, an Exadata system will offload the scanning of the table to the Exadata storage,
filter out all sales orders that are less than $1000, filter out sales orders not in March, and extract just
the relevant customer information. This reduces the data transferred to the database servers by orders
of magnitude. Smart Scan greatly accelerates query execution, eliminates bottlenecks, and “[With Exadata] We can more
significantly reduces the CPU usage of the database servers. quickly process 65 billion daily
transactions for data charging,
while providing real-time
Storage Index is another powerful unique capability of Oracle Exadata System software that helps
information for customer
avoid unnecessary I/O operations and improves overall performance. The storage index, maintained
inquiries, increasing customer
in-memory at the storage server, tracks summary information for table columns contained in a storage satisfaction, and reducing
region on that storage server. When a query specifies a WHERE clause, Exadata System software costs.”
examines the storage index using a Bloom filter to determine if rows with the specified column value
Jin Hyung Lee
might exist in a region of disk on the storage server. If the column value doesn’t exist in the Bloom
ICT Team Manager, Networking
filter, then scan I/O in that region for that query is skipped. Storage Indexes make many SQL Engineering
operations run dramatically faster because large numbers of I/O operations are automatically replaced SK Telecom
by a few in-memory lookups.
Besides the intrinsic capabilities of Exadata System software, the combination of Oracle Database
software, Exadata System software and Exadata infrastructure enables several additional unique
capabilities that offer unparalleled performance levels for OLTP workloads. For example, Exafusion
Direct-to-Wire Protocol uniquely allows database processes to read and send Oracle Real
Applications Cluster (Oracle RAC) messages directly over the ultra-fast RoCE network using Remote
Direct Memory Access (RDMA), bypassing the OS kernel and networking software overhead. This
improves the response time and scalability of Oracle RAC OLTP configurations on Oracle Exadata
Database Machine, especially for workloads with high-contention updates.
In some OLTP workloads, more than half of remote reads are for Undo Blocks to satisfy read
consistency. Exadata uniquely leverages ultra-fast RDMA to read UNDO blocks from other database
instances, further improving OLTP performance.
In addition, Exadata uniquely uses Machine Learning to implement Automatic Indexing with Oracle
Database 19c. Automatic Indexing continually analyzes executing SQL and creates new indexes to
accelerate performance. Automatic Indexing continuously learns and tunes the database as the
underlying data model or usage patterns change.
Exadata also uniquely implements Real Time Statistics gathering as DML operations insert, update
or delete data. Real Time Statistics allows the SQL optimizer to adapt plans dynamically as the
distribution of data changes.
Two modes of Hybrid Columnar Compression are available. Warehouse compression mode is
suitable for read-intensive workloads and provides large storage savings and enhanced analytic
performance. Archive compression mode provides the highest degree of compression and is
targeted at data that is seldom accessed but still must be kept online. In addition, this data can now be
seamlessly stored on the XT storage server for further cost reduction.
On OLTP systems, Hybrid Columnar Compression can be used to compress older, less active data
while newer, more active and update-intensive data can be compressed using Advanced Row
Compression. Oracle Database Release 18c and above provides the ability to change the type of
compression used by individual table partitions online (even if there are global indexes on the table), to
ensure seamless tiering across different compression types as data ages and becomes less active.
Exadata uniquely implements Fault Tolerant memory duplication for Oracle Database In-Memory.
On a generic cluster configuration, when a server node fails, the in-memory data on that node is lost,
and it takes many minutes to repopulate the in-memory data on a surviving node. During this time,
analytic queries will run orders of magnitude slower. This means generic platforms will fail to meet “Our critical electronic
business SLAs. However, on Exadata, Fault-Tolerant memory duplication can eliminate this slowdown payments service has been live
by duplicating any subset of the in-memory data across the clustered database servers. If a database on Exadata since early 2011
with 100% uptime. The service
server fails, queries will transparently access the duplicate copy on a surviving database server and
reliably processes the transfer
continue without interruption. of billions of Euros per week
and achieves subsecond
Exadata uniquely integrates with Active Data Guard to allow customers to run In-Memory analytics on response times for online
a standby database, further improving the return on investment of the standby system, and enhancing enquiries.”
availability and overall performance.
Martin McGeough
Database Technical Architect
ENHANCED CONSOLIDATION USING ORACLE VIRTUAL MACHINES Vocalink
Consolidated environments running on Exadata X8M use KVM-based Oracle Virtual Machines
(OVMs) for a high degree of isolation between workloads, especially in hosted, shared, service
provider, and test/dev environments. Using OVM, multiple RAC clusters can be deployed on the same
set of database servers of the Exadata Database Machine, for consolidation of applications with
specific Clusterware version needs.
Exadata Database Machine is the world’s fastest virtualized database platform. Exadata virtual
machines use high speed networking with Single Root I/O Virtualization (SR-IOV) to ensure that
performance within a virtual machine is similar to Exadata’s excellent raw hardware performance.
Virtual machines on Exadata are considered Trusted Partitions and therefore software can be licensed
at the virtual machine level instead of the physical processor level. Without Trusted Partitions,
database options and other Oracle software must be licensed at a server or cluster level even though
“By consolidating 350 database
not all databases running on that server or cluster may require a particular option.
servers and storage systems
onto Oracle Exadata, we gained
ENTERPRISE-CLASS SECURITY WITH EXTREME PERFORMANCE a high-performance, reliable,
and scalable mobile billing
Exadata Database Machine is the world's most secure database machine. Building on the high platform, enabling us to
security capabilities in the Oracle Database such as Transparent Data Encryption (TDE), Exadata calculate billings data 10x
uniquely moves decryption processing from database server software into the Exadata Storage faster, and halve maintenance
Server hardware. Exadata storage leverages hardware decryption and compression together to costs.”
provide the highest performance secure databases. Encryption occurs after the data is Tomoki Shimamura
compressed so that the cost of decryption is decreased by the degree of compression. By Senior Manager Billing Systems
leveraging both technologies, Exadata is able to query fully encrypted and compressed databases with Group
minimal overhead at a rate of hundreds of gigabytes of (original) user data per second. In addition, NTT DoCoMo, Inc.
TDE provides a complete key management solution to keep all data encrypted and secure.
Exadata security has been probed and evaluated by hundreds of leading banks, telecoms, and
government organizations worldwide. The security findings of all these evaluations have been
incorporated into the Exadata standard configuration. Therefore Exadata benefits from scrutiny both
by Oracle Security experts and by hundreds of industry security experts around the world.
configuration to transparently maintain a real-time copy of the database at a remote site and provide Roland Schiller
full protection against primary database failures and site disasters. Architect
Swiss Re
Exadata in an MAA configuration is recognized by the analyst firm IDC as a system that delivers at
least 5-nines availability and is categorized in the IDC AL4 fault-tolerant market segment, along with
HP Integrity NonStop and IBM z Systems1.
The Exadata principle of deep hardware and software integration is also evident in the many ways
Exadata uniquely assures high availability across several different failure conditions. One such unique
capability is Instant Detection of Compute and Storage Server Failures. On non-Exadata
platforms, detecting a server failure requires waiting for a long timeout, leading to extended application
brownouts. Exadata X8M implements a unique RDMA-based sub-second node death detection,
leading to virtual elimination of application brownout conditions.
Disk and flash devices occasionally exhibit very long latency I/O operations due to internal recovery of
failed sectors, internal firmware reboots, or wear leveling. These long I/O operations can cause stalls
in mission critical OLTP databases. With Exadata’s unique I/O Latency Capping, Oracle Exadata “We’ve seen system
System software automatically redirects read I/O operations to an ASM-mirrored copy of the data performance increase 4x and
the total cost of storage
when the latency of a read I/O is much longer than expected. Similarly, it automatically redirects high
ownership fall by 30%, even as
latency write I/O operations to a healthy flash device, eliminating write outliers. Exadata System our data volumes have
Software uses Machine Learning techniques to predict components susceptible to failure and takes surged.”
proactive action to gracefully take such components out of service. If disks do fail, ASM performs a
Holger Haun
rebalance operation for the data that was resident on the disk. Exadata allows hot swapping of disks,
Managing Director
fans, power supplies, and PCIe Flash cards to avoid downtime. Exadata System software takes IDS GmbH – Analysis and
rebalance one step further by preserving the flash cache population and storage indexes when moving Reporting Service
data between storage servers to maintain consistent application performance. On rare occasions
when there are outliers within the networking subsystem, Exadata redirects the I/O issued by the
database server to another storage server.
Exadata automates monitoring of CPU, memory, Input/Output, file system, and network. This
automation combines machine learning techniques with the deep lessons learned from thousands of
mission critical real-world deployments. For example, Exadata can detect that anomalous use of
system resources is affecting database performance, identify the process responsible, and issue an
alert, without any pre-existing set-up.
Because of its industry leading availability, the Exadata Database Machine has been deployed by
leading companies for their most critical applications including interbank fund transfers, online
securities trading, real-time call tracking, and web-based retailing. Exadata’s Mission Critical
1
Worldwide Fault-Tolerant Servers Market Shares, 2014: Vendors Are Hearing the Customer – More Bold Moves Needed to Grow the Segment, IDC,
Peter Rutten, Lloyd Cohen, October 2015
Exadata uniquely implements database and I/O resource management. Fine-grained priorities
specified for operations at the database level are automatically communicated to Exadata Storage
Servers and applied to each I/O operation to ensure that prioritization of database operations applies
to both CPU operations and I/O operations. The same resource management principles can also be
applied when multiple databases are deployed on one Exadata rack, as is typical in a consolidated
private cloud.
New in the X8M generation, Exadata utilizes RDMA over Converged Ethernet protocols to ensure
network intensive workloads such as reporting, batch, and backups don’t stall latency sensitive
interactive workloads. Latency sensitive network operations, such as RAC Cache Fusion
communication, and log file writes, travel across high priority network channels within the converged
ethernet fabric. Non-latency sensitive traffic travels on other channels, with their own network switch
buffers.
Because of Exadata’s unique Consolidation and Database as a Service capabilities, Exadata is the
only platform that can support up to 4000 Pluggable Databases within a single Oracle Multitenant
Container Database.
Snapshots start with a shared read-only copy of the production database (or PDB) that has been
cleansed of sensitive information. A hierarchy of read-write snapshots can be created from this shared
copy. As changes are made, each snapshot writes the changed blocks to a sparse disk group. Since
multiple users can create independent snapshots from the same base database copy, multiple test
and development environments can share space while maintaining independent databases for each
task.
In addition, Exadata supports Docker containers, enabling customers to easily and rapidly provision
databases for testing and development to foster agile development. Customer can also use Docker
containers to host lightweight applications and agents.
• Unified Monitoring: Oracle Enterprise Manager 13c uniquely supports a single pane of glass
view of all the hardware and software components such as database servers, storage
servers, network switches, and monitors the operations running on them and their resource
utilization. DBAs can drill down from database monitoring screens to the Exadata storage
layer to quickly determine root causes of any performance bottlenecks.
• Lights-out monitoring within Enterprise Manager is optimized for Exadata with predefined
metrics and thresholds so that administrators receive timely notifications when issues arise,
and manage those exceptions. In addition, hardware incidents are automatically detected and
service requests logged to reduce problem resolution time.
• The Exachk tool, which is integrated with Enterprise Manager’s powerful compliance
framework, provides functionality for system administrators to automate the assessment of
Engineered Systems for known configuration problems and best practices. Administrators
can leverage the Consistency Check functionality to check for deviations in configuration
across the racks or among the database servers of a rack.
• Exadata’s built in Management Server (MS) processes constantly monitor the health of
hardware and software components, and send alerts to both administrators and Oracle
support when faulty components are detected.
Of particular value is Oracle Platinum Services that is available exclusively for Oracle’s Engineered
Systems. Platinum Services provides fault monitoring, faster response times, and expedited escalation
to development. With Platinum Services, Oracle support engineers perform software maintenance and
patching remotely. Platinum Services provides a higher level of support than has ever been available
before for all software and hardware within an Engineered System including the Oracle Database.
Platinum Services is provided at no extra charge to Exadata customers.
IT AGILITY
Exadata is a complete system for running databases including storage, servers, and internal networks.
Management of a traditional database system is typically spread across the management teams of
each of the components such as the database team, the storage team, and the system administration
team. In contrast, an Exadata system is typically managed by a unified Database Machine
Administration (DMA) team. Database Machine Administrators have full control of all resources in
the Exadata Database Machine including storage resources. New database deployments and
Exadata provides a huge RAM, flash, and disk footprint for large data sets. Raw disk storage on an
Exadata full rack can exceed 3 Petabytes while raw flash storage can be up to 920 TB. In addition,
Hybrid Columnar Compression often expands storage and memory capacity 10X. By intelligently
moving active data across disk, flash, and memory tiers, Exadata simultaneously delivers the highest
performance and the lowest cost.
Exadata has the unique ability to consolidate many databases supporting multiple workloads in a
single cloud platform. High-end OLTP, analytics, batch, reporting, and backups can all run
simultaneously within and across databases with extreme performance. The extreme performance
and capacity of Exadata enables very large numbers of databases and workloads to be
consolidated on Exadata. Consolidating databases on Exadata reduces system hardware cost,
software cost, and greatly reduces ongoing operations cost.
The uniformity of Exadata Database Machine configurations results in large cost savings. Exadata
standardizes not just technologies, but also integration, testing, security, hardening, tuning,
and support. Customers deploy Exadata systems much faster and with a lot less labor than traditional
systems. Low level tuning, integration, and maintenance is reduced or eliminated. Because all
Exadata users run a configuration that is identical to thousands of other users, and is identical to
Oracle’s internal configurations, it is far less likely that issues will be encountered, and issue resolution
is quicker and simpler reducing both operations cost and downtime cost.
Oracle databases deployed in the cloud as part of this service include all Oracle Database options
and features – such as Oracle Multitenant, In-Memory Database, Real Application Clusters (RAC),
Active Data Guard, Partitioning, Advanced Compression, Advanced Security, etc. It also includes all
• Running business-critical production OLTP or analytic databases at almost any scale without
incurring the capital expenditure and complexity of maintaining the underlying IT
infrastructure.
• Consolidating a variety of workloads in the Cloud using multiple Oracle Databases or Oracle
Multitenant.
• Maintaining synchronized Oracle standby or replica databases for disaster recovery in the
Cloud and/or query offloading using Oracle Active Data Guard or Oracle GoldenGate.
• Quickly provisioning high-performance Oracle databases for ad-hoc business reasons such
as feature development, functionality testing, application certification, proof-of-concept, try-
before-buy, Database sandbox testing, etc.
• Executing time-sensitive business applications such as launching a web-based marketing
campaign, running loyalty programs, rolling out new business initiatives, etc.
An attractive aspect of all these use cases for existing Oracle Database customers is that their
applications and data models do not have to change. Their data center simply expands to include the
elasticity and flexibility of the Oracle Cloud. They also do not have to invest in multiple cloud platforms
for multiple workloads since Exadata provides a unified platform for all workloads – analytics, data
warehousing, OLTP, consolidation, in-memory, and mixed-workloads.
With a database platform uniquely engineered for extreme performance for all workloads, along with
fast deployment, simplified management, low operating costs and reduced risks, Exadata Cloud
Service is the best Cloud Database platform available today.
Customers that are not able to move their databases to the Public Cloud can get the Oracle Public
Cloud implemented in their own data center using the Oracle Database Exadata Cloud at Customer.
Exadata Cloud at Customer:
• Is paid for using subscription based pricing that includes all Database and Exadata software,
all Exadata hardware, and remote operation and infrastructure management by Oracle.
• Includes the same Web and REST based DB and OS provisioning, management, and
orchestration as the Oracle Public Cloud.
• The Oracle Cloud Operations Team Manages the Exadata Infrastructure including Servers,
storage, storage software, networking, firmware, hypervisor, etc.
• Exadata is the only platform suitable for deploying Mission Critical databases using Cloud at
Customer.
Exadata accelerates time to market for new business applications since the time needed for system
configuration, tuning, and testing is largely eliminated. Deployment times are reduced from months to
days, and the risk of unexpected system level issues after go-live is greatly reduced. When a new
application is deployed, it is common for unanticipated application usage patterns to create
performance issues. Exadata’s huge I/O, network, and compute throughput can absorb spikes created
Exadata’s extreme performance, large memory, and flash capacity enhance employee productivity
and customer satisfaction by greatly improving user response times. Users spend more time doing
useful work, and less time waiting for the system to respond.
Exadata’s extreme performance does not just improve business efficiency, it also enables business
users to make smarter decisions, discover growth opportunities, and reduce costs. Users can
analyze data in real-time, explore different possibilities, and perform rapid iteration to find better
solutions. Exadata enables:
CONCLUSION
Exadata delivers a fully integrated database platform with the latest hardware technologies and
unique software to deliver extreme performance, availability, and security. This coupled with cost
savings, ease of management, and enhanced supportability result in greater business agility and
efficiency. Given what can be achieved with Exadata, it is no surprise it is the new global standard for
running Oracle Databases – whether on-premises, or in the Oracle Cloud.
Database Server 2x 24-core Intel Xeon 8260 384 GB (default) to 1.5 TB 4x 1.2 TB None 2x 10 Gb copper Ethernet ports, or
processors (2.4GHz) (max) 10,000 RPM
2x 10/25 Gb optical Ethernet ports (client)
disks (hot-
2x 10/25 Gb optical Ethernet
swappable)
4x 10 Gb copper Ethernet ports, or
2x 10/25 Gb optical Ethernet (optional)
1
All servers include redundant hot swappable fans and power supplies
2
Table includes individually purchasable servers only, Eighth Rack Database and EF Storage Servers also available in rack configuration shown below. HC = High Capacity, EF =
Extreme Flash, XT = Extended.
RACK SIZE DATABASE SERVERS STORAGE SERVERS AND HIGH CAPACITY STORAGE EXTREME FLASH
AND CORES CORES SERVER CAPACITY (RAW) STORAGE SERVER
CAPACITY (RAW)
Eighth Rack3 2x servers, 48 cores 3x servers, 48 cores for SQL 252 TB disk, 38.4 TB Flash, 4.5 76.8 TB Flash, 4.5 TB
offload TB Persistent Memory Persistent Memory
Quarter Rack 2x servers, 96 cores 3x servers, 96 cores for SQL 504 TB disk, 76.8 TB Flash, 4.5 153.6 TB Flash, 4.5 TB
offload TB Persistent Memory Persistent Memory
Half Rack4 4x servers, 192 cores 7x servers, 224 cores for SQL 1176 TB disk, 179.2 TB Flash, 358.4 TB Flash, 10.5 TB
offload 10.5 TB Persistent Memory Persistent Memory
Full Rack4 8x servers, 384 cores 14x servers, 448 cores for SQL 2,352 TB disk, 358.4 TB Flash, 21 or 716.8 TB Flash, 21 TB
offload TB Persistent Memory Persistent Memory
+Storage Servers n/a Up to 18x servers5, 576 cores 3,024 TB disk, 460.8 TB Flash, 921.6 TB Flash, 27.6 TB
max per rack 27.6 TB Persistent Memory Persistent Memory
maximum per rack maximum per rack
Multi-Rack Connection Connect any combination of up to 18 Exadata Database Machine racks or Exadata Storage Expansion Racks via the RoCE
Network Fabric. Larger configurations can be built with external RoCE switches. Connected racks must contain Exadata RoCE
hardware.
Eighth Rack Expansion Options Expand just compute or just storage or both, described as follows:
• Database Servers can be expanded by installing an additional 24 core CPU per server
• EF Storage Servers can be expanded by enabling 16 cores and installing 4 PCI Flash cards per server
• HC Storage Servers can be expanded by adding Eighth Rack High Capacity Storage Server
SERVER TYPE MAXIMUM SQL MAXIMUM SQL MAXIMUM SQL PERSISTENT PCI FLASH DISK DATA
FLASH READ IOPS1,3 WRITE IOPS4 MEMORY CAPACITY CAPACITY (RAW)
BANDWIDTH2 CAPACITY (RAW) 5 (RAW)5
1
HC = High Capacity, EF = Extreme Flash, XT = Extended. PMEM = Persistent Memory. Actual system performance varies by application.
2
Bandwidth is peak physical scan bandwidth achieved running SQL, assuming no database compression. Effective user data bandwidth is higher when database compression is
used.
3
Based on 8K I/O requests running SQL. Note that the I/O size greatly affects Flash IOPS. Other products quote IOPS based on smaller I/Os that are not relevant for databases.
4
Based on 8K I/O requests running SQL. Flash write I/Os measured at the storage servers after ASM mirroring, which usually issues multiple storage I/Os to maintain redundancy.
5
Raw capacity is measured in standard disk drive terminology with 1 GB = 1 billion bytes.
FLASH METRICS MAXIMUM SQL FLASH MAXIMUM SQL PMEM MAXIMUM SQL FLASH PCI FLASH CAPACITY
BANDWIDTH2 READ IOPS1,3 WRITE IOPS4 (RAW)5
1
EF = Extreme Flash; HC = High Capacity; PMEM = Persistent Memory
2
Bandwidth is peak physical scan bandwidth achieved running SQL, assuming no database compression. Effective user data bandwidth is higher when database compression is
used.
3
Based on 8K IO requests running SQL. Note that the IO size greatly affects Flash IOPS. Others quote IOPS based on smaller IOs and are not relevant for databases.
4
Based on 8K IO requests running SQL. Flash write I/Os measured at the storage servers after ASM mirroring, which usually issues multiple storage IOs to maintain redundancy.
5
Raw capacity is measured in standard disk drive terminology with 1 GB = 1 billion bytes. Usable capacity is measured using normal powers of 2 space terminology with 1 TB = 1024
* 1024 * 1024 * 1024 bytes.
Exadata Typical Rack Configurations: Disk Capacity and Performance Metrics (HC)
DISK METRICS MAXIMUM SQL DISK BANDWIDTH1 MAXIMUM SQL DISK IOPS2 DATA CAPACITY (RAW)3
1
Bandwidth is peak physical scan bandwidth achieved running SQL, assuming no database compression. Effective user data bandwidth is higher when database compression is
used.
2
Based on 8K IO requests running SQL. Note that the IO size greatly affects Flash IOPS. Others quote IOPS based on smaller IOs and are not relevant for databases.
3
Raw capacity is measured in standard disk drive terminology with 1 GB = 1 billion bytes. Usable capacity is measured using normal powers of 2 space terminology with 1 TB = 1024
* 1024 * 1024 * 1024 bytes.
DATA CAPACITY (USABLE) – NORMAL DATA CAPACITY (USABLE) – HIGH MAXIMUM DATA LOAD
COMBINED METRICS REDUNDANCY1 REDUNDANCY1 RATE2
EF 56 TB 44 TB 7.5 TB/hour
EF 28 TB 22 TB 3.8 TB/hour
1
Usable capacity is measured using normal powers of 2 space terminology with 1 TB = 1024 * 1024 * 1024 * 1024 bytes. It is the actual space available to create a database after
taking into account space needed for ASM redundancy, recovering from a drive failure. Normal redundancy calculations reflect the use of Grid Infrastructure version 12.2.0.1 or later.
2
Load rates are typically limited by database server CPU, not I/O. Rates vary based on load method, indexes, data types, compression, and partitioning.
Weight 45.6 lb (20.7 kg) 76.7 lb (34.8 kg) 60.6 lb (27.5 kg) 66.7 lb (30.2 kg) 67.5 lb (30.6 kg)
Maximum power usage 0.7 kW (0.7 kVA) 0.8 kW (0.8 kVA) 0.8 kW (0.8 kVA) 0.5 kW (0.5 kVA) 0.6 kW (0.6 kVA)
Typical power usage1 0.5 kW (0.5 kVA) 0.5 kW (0.6 kVA) 0.6 kW (0.6 kVA) 0.3 kW (0.3 kVA) 0.4 kW (0.4 kVA)
Cooling at maximum usage 2,409 BTU/hour 2,631 BTU/hour 2,730 BTU/hour 1,570 BTU/hour 1,947 BTU/hour
2,541 kJ/hour 2,775 kJ/hour 2,880 kJ/hour 1,656 kJ/hour 2,054 kJ/hour
Cooling at typical usage 1,686 BTU/hour 1,842 BTU/hour 1,911 BTU/hour 1,099 BTU/hour 1,363 BTU/hour
1,779 kJ/hour 1,943 kJ/hour 2,016 kJ/hour 1,159 kJ/hour 1,438 kJ/hour
Airflow at maximum usage2 112 CFM 122 CFM 126 CFM 73 CFM 90 CFM
Weight 2016.70 lb (914.8 kg) 1296.4 lb (588.0 kg) 907.4 lb (411.6 kg) 879.9 lb (399.1 kg)
Maximum power usage 17.3 kW (17.6 kVA) 9.1 kW (9.3 kVA) 4.6 kW (4.7 kVA) 3.7 kW (3.8 kVA)
Typical power usage 1 12.1 kW (12.3 kVA) 6.4 kW (6.5 kVA) 3.2 kW (3.3 kVA) 2.6 kW (2.6 kVA)
Cooling at maximum usage 59,010 BTU/hour 30,958 BTU/hour 15,617 BTU/hour 12,602 BTU/hour
Cooling at typical usage 41,307 BTU/hour 21,671 BTU/hour 10,932 BTU/hour 8,821 BTU/hour
Airflow at maximum usage 2 2732 CFM 1433 CFM 723 CFM 583 CFM
Airflow at typical usage 2 1912 CFM 1003 CFM 506 CFM 408 CFM
Weight 1791.3 lb (812.5 kg) 1183.7 lb (536.9 kg) 859.1 lb (389.7 kg) 854.0 lb (387.4 kg)
Maximum power usage 17.7 kW (18.1 kVA) 9.3 kW (9.5 kVA) 4.7 kW (4.8 kVA) 3.8 kW (3.9 kVA)
Typical power usage 1 12.4 kW (12.6 kVA) 6.5 kW (6.6 kVA) 3.3 kW (3.3 kVA) 2.6 kW (2.7 kVA)
Cooling at maximum usage 60,395 BTU/hour 31,651 BTU/hour 15,914 BTU/hour 12,903 BTU/hour
Cooling at typical usage 42,276 BTU/hour 22,156 BTU/hour 11,140 BTU/hour 9,032 BTU/hour
Airflow at maximum usage 2 2796 CFM 1465 CFM 737 CFM 597 CFM
Airflow at typical usage 2 1957 CFM 1026 CFM 516 CFM 418 CFM
Operating temperature/humidity: 5 ºC to 32 ºC (41 ºF to 89.6 ºF), 10% to 90% relative humidity, non-condensing. Altitude Operating: Up to 3,048 m, max. ambient temperature is de-
rated by 1° C per 300 m above 900 m
1
Typical power usage varies by application load. 2 Airflow must be front-to-back.
Certifications 2,3 North America (NRTL), European Union (EU), International CB Scheme, HSE Exemption (India), BSMI
(Taiwan), CCC (PRC), EAC (EAEU including Russia), RCM (Australia), VCCI (Japan), Mexico, KC (Korea)
European Union Directives 3 2014/35/EU Low Voltage Directive, 2014/30/EU EMC Directive, 2011/65/EU RoHS Directive, 2012/19/EU
WEEE Directive
1
All standards and certifications referenced are to the latest official version at the time the data sheet was written. For additional detail, please contact your sales representative.
2
Other country regulations/certifications may apply.
3
In some cases, as applicable, regulatory and certification compliance were obtained for the shelf-level systems only.
• Hardware Warranty: 1 year with a 4 hr web/phone response during normal business hours (Mon-Fri 8AM-5PM), with 2 business day on-site response/Parts
Exchange
• Oracle Premier Support for Systems includes Oracle Linux support and 24x7 with 2 hour on-site hardware service response (subject to proximity to service
center)
• Oracle Premier Support for Operating Systems
• Oracle Customer Data and Device Retention
• System Installation Services
• Software Configuration Services
• Oracle Platinum Services
• Business Critical Service for Systems
• Oracle Exadata Start-Up Pack
• System Upgrade Support Services including hardware installation and software configuration
• Oracle Auto Service Request (ASR)
Optional Customer Supplied Ethernet Switch Installation in Exadata Database Machine X8M-2
Each Exadata Database Machine X8M-2 rack has 2U available at the top of the rack that can be used by customers to optionally install their own client network Ethernet
switches in the Exadata rack instead of in a separate rack. Some space, power, and cooling restrictions apply.
Manageability Features
• Oracle Embedded Integrated Lights Out Manager (ILOM)
• Oracle Enterprise Manager Exadata Plug-in
• Unique Active AWR includes storage stats for end to end monitoring
• IPv6 Support for Ethernet Connections
• Capacity on Demand
• Cell software transparent restart
• Flash and disk life cycle management alert
• Automatic Disk Scrub and Repair
• Trusted Partitions for Oracle Virtual Machine
• Automated VLAN Creation
• Oracle Exadata Deployment Assistant
• Separate Management Switch and Connectivity
• Exacli command line management from remote servers
• Cellcli command line management of Storage Servers
CONNECT WITH US
Call +1.800.ORACLE1 or visit oracle.com.
Outside North America, find your local office at oracle.com/contact.
Copyright © 2020, Oracle and/or its affiliates. All rights reserved. This document is provided for information purposes only, and the contents hereof are
subject to change without notice. This document is not warranted to be error-free, nor subject to any other warranties or conditions, whether expressed
orally or implied in law, including implied warranties and conditions of merchantability or fitness for a particular purpose. We specifically disclaim any
liability with respect to this document, and no contractual obligations are formed either directly or indirectly by this document. This document may not be
reproduced or transmitted in any form or by any means, electronic or mechanical, for any purpose, without our prior written permission.
Oracle and Java are registered trademarks of Oracle and/or its affiliates. Other names may be trademarks of their respective owners.
Intel, Intel Optane and Intel Xeon are trademarks or registered trademarks of Intel Corporation. All SPARC trademarks are used under license and are
trademarks or registered trademarks of SPARC International, Inc. AMD, Opteron, the AMD logo, and the AMD Opteron logo are trademarks or registered
trademarks of Advanced Micro Devices. UNIX is a registered trademark of The Open Group. 0120