Welcome to the Fujitsu M10 server Features and Capabilities session.
This course will present Fujitsu M10 Server hardware features as well as software information about Fujitsu M10 virtualization options.
After completing this session, you will be able to:
Describe the Fujitsu M10 Server product lineup
Define the hardware features of Fujitsu M10 Servers
Describe the RAS features of Fujitsu M10 Servers
Describe the Fujitsu M10 Servers Functions and Features
Describe the Virtualization options of Fujitsu M10 Servers
Here is a high-level overview of the Oracle product portfolio.
Oracle's server portfolio consists of four server lines. Let’s focus on the SPARC servers running Oracle Solaris. The SPARC server line is the ideal platform for mission-critical applications that require high performance, scalability and best-in-class availability.
In early 2013, the Fujitsu M10 system was added into the SPARC server portfolio. The Fujitsu M10 system continues with the SPARC64 processor family and XSCF service processor management from the previous SPARC Enterprise M-Series. However, Fujitsu M10 servers feature many unique characteristics such as dynamic scaling with COD and a modular architecture, Software on Chip (SWoC), and enhanced capacity up to 1,024 cores. The Fujitsu M10, together with the SPARC servers developed by Oracle, comprise a comprehensive portfolio for mission-critical computing.
Here is the Fujitsu M10 product lineup, which covers a wide range of options.
The M10-1 server is a 1 processor, 1RU space-saving, entry-level server, and yet it offers all of the high performance and reliability of a SPARC Enterprise M-Series server. It can be configured with up to 16 CPU cores, for scalability, and is best for integration and/or virtualization of horizontally-scaled data deployments.
The M10-4 server is a 4 processor, 4RU mid-range, rackmount server with high performance and reliability. With scalability up to 64 CPU cores and 2TB of memory, it offers immense processing capability in a single box.
The M10-4S server adds flexible scalability, high performance and high reliability. It is the best server for conventional computing workloads and for IT infrastructures that support new compute paradigms, such as cloud computing and big data processing.
The M10-4S is a 4RU, 4 processor building block that can be connected together, like stacking Lego blocks, to create single instance servers from 4 to 64 processors. The system can be expanded in increments from 1 to 16 units, all the way up to 1024 processor cores and 32TB of memory. Chassis are connected by Fujitsu’s unique interconnect technology that includes leading supercomputing innovations to ensure high-bandwidth and linear scalability. It is important to note that M10-4S Building Blocks are not interconnected in a cluster of individual Solaris instances but when connected can create a single, large shared memory Solaris instance. The interconnect is transparent to the Solaris operating system.
The latest SPARC64 X (ten) and X+ (ten plus) processors are available in all three models, so customers can enjoy the benefits of Software on Chip and core-level CPU Activation, known as COD. The Fujitsu M10-4S supports mixed SPARC64 X (ten) and X+ (ten plus) building blocks in a single system.
All models support Oracle Solaris 11 and 10, as well as Oracle Solaris Zones, providing the flexibility to run four generations of Oracle Solaris.
M10-4S implements a unique modular approach to scale from 2 or 4 sockets all the way to 64 sockets.
This is called Building Block architecture. Each M10-4S 4RU chassis is called a Building Block or BB for short.
2, 3, and 4 BB configurations can be interconnected with Expansion kits that include copper interconnect cables. Up to 4 interconnected BBs can be installed in standard 19inch racks.
Configurations with more than four M10-4S Building Blocks require a first expansion rack that includes optical interconnect cables and crossbar boxes.
Configurations with more than eight M10-4S Building Blocks require a second expansion rack that also includes optical interconnect cables and two more crossbar boxes.
If an M10-4S configuration is planned to grow larger than 4 BBs, it is recommended to start with the first expansion rack. This will prevent the need to re-cable when switching from copper to optical interconnect cables.
First off, we can assume that customers are always looking to achieve better optimization of resources for key mission-critical applications. However, nobody can predict the future growth of their business with perfect accuracy. So, customers face a dilemma: A small system might be able to fit into the customer's current business, but future business growth will render it useless. How about using a large system? Costly resources will be wasted until the customer's business grows to expectations. It is difficult for customers to find the optimal system.
In addition, customers who are considering efficient server consolidation face the complexity of managing too many individual systems running mission-critical workloads that demand high levels of RAS and guaranteed data integrity. Without a high level of RAS, a single error can cause critical problems for multiple applications running on a consolidated, single system.
Finally, business moves and grows in ever-changing ways, and customers are looking for systems that will ease the decision-making process and lead them in the right direction sooner than their competitors.
Fujitsu M10 servers are optimal for all of these situations, because they provide unique dynamic scaling, mission-critical RAS and extreme performance for a wide range of enterprise applications, including OLTP, Business Intelligence, Data Warehousing and business applications. This is due to Fujitsu M10’s combination of multiple leading-edge technologies.
All M10 systems enable processor COD on a core-by-core basis via CPU activation. Fujitsu M10-4S systems provide further and unique scalability to grow from a small 4-socket server up to a large 64-socket system with a modular architecture. This unique dynamic scalability provides customers with M10’s first key value: “Pay as You Grow.”
Customers can optimize their resources on as small a basis as one core through CPU activation, and add CPUs, or whole servers with the Building Block architecture using Fujitsu’s unique high-speed interconnection technology. This allows linear performance improvement up to the maximum 16-Building Block, 64-socket, 1,024-core configuration. This flexibility, coupled with the industry’s leading scalability, will allow customers to adapt to any size workload, and is ideal for big data processing infrastructures that start on a large scale or may grow significantly over time. Customers can leverage the three no-cost virtualization technologies of physical partitioning, Oracle VM for SPARC, and Solaris Zones. Customers can also mix Building Blocks using different generations of processors, which protects their valuable investment.
Being the "Most Trusted" is the second key value. Fujitsu M10 systems not only inherit the RAS features of the previous M-Series, but also improve on them to provide guaranteed data integrity. Key components in the Fujitsu M10 system are hot-swappable and redundantly configurable. In addition, error checking and self-healing features, honed from decades of mainframe experience, are implemented in all circuits of the SPARC64 X (ten) and X+ (ten plus) processors. This mainframe-class reliability makes Fujitsu M10 an ideal platform for "always-on" operations that maximize opportunity gains for customers.
The third key value is "real-time decisions" supported by industry leading performance. You can see the innovative Software on Chip technology in all Fujitsu M10 systems. Engineers at Oracle and Fujitsu worked together on the OS, Database, and Compiler to implement common software operations in the processor hardware, providing extreme high performance for Solaris applications. Combining huge, high-speed memory and core capacity with proven Solaris scalability, M10 systems are ideal for real-time processing and mission critical applications in this age of Big Data.
Here is an overview of the SPARC64 X (ten) and X+ (ten plus) processors.
They run at a high CPU clock speed and implement a 64 socket high speed scalable interconnect.
SPARC64 X (ten) runs at up to 3.0GHz and its CPU-to-CPU interconnect speed is 14.5Gbps.
SPARC64 X+ (ten plus) runs at up to 3.7GHz and its CPU-to-CPU interconnect speed is 25.0Gbps.
Interconnect speed is per lane and there are 8 unidirectional lanes per interconnect.
Each socket contains four Memory Access Controllers, or MACs, that support the latest 1,600MHz or 1,333MHz DDR3 DIMMs. Each MAC has two memory buses that connect the CPUs and DIMMs directly.
No Buffer-on-Board or BOB is required. This achieves a 100ns load-to-use latency.
Two PCI Express Generation 3 controllers are also integrated in each socket and provide two 8-lane PCI Express Generation 3 ports.
Small detail, but having 2 controllers in the chip gives us full IO path redundancy, even in the 1 socket M10-1.
L2 cache is 24MB shared by 16 cores or 32 threads.
Each core contains two threads that implement Simultaneous Multi-Threading, or SMT, technology.
L1 cache in the core consists of instruction cache, or “I-Cache,” and data cache, or “D-Cache.”
Each cache size is 64KB.
SPARC64 X (ten) and X+ (ten plus) implement the sun4v architecture.
In addition to ECC and parity protection, both SPARC64 X (ten) and X+ (ten plus) implement
Fujitsu's heritage of mainframe-class RAS, such as Hardware Instruction Retry and
Processor History logging.
CPU power management like Dynamic Voltage and Frequency Scaling, or DVFS,
can be used to reduce power consumption.
This table shows an overview of the Fujitsu M10 servers.
All Fujitsu M10 models are designed to take full advantage of the exceptional power and performance of SPARC64 X (ten) and X+ (ten plus) processors.
SPARC64 X+ (ten plus) runs at up to 3.7GHz, and SPARC64 X (ten) runs at up to 3.0GHz.
They provide 16 cores per processor with 2 SMT threads per core.
The Fujitsu M10-1 is a 1RU rack mount server that contains one 3.2GHz SPARC64 X+ (ten plus) or
one 2.8GHz SPARC64 X (ten) processor that can be extended from a minimum of 2 cores to a maximum of 16 cores in steps, using CPU activation.
It has 16 memory slots for up to 512MB of memory and 3 PCI Express Generation 3 slots.
The Fujitsu M10-4 is a 4RU rack mount server that can contain two or four 3.4GHz SPARC64 X+ (ten plus) or
2.8GHz SPARC64 X (ten) processors that can be extended from a minimum of 4 cores
to a maximum of 64 cores in steps, using CPU activation. M10-4 servers can be expanded in the field from 2 to 4 CPU sockets.
They have 64 memory slots for up to 2TB of memory and 11 PCI Express Generation 3 slots.
The Fujitsu M10-4S is a 4RU rack mount server that can contain two or four 3.7GHz SPARC64 X+ (ten plus) or
3.0GHz SPARC64 X (ten) processors and is scalable by stacking up chassis (known as Building Blocks or BBs).
Each BB can be extended from a minimum of 4 cores to max 64 cores in steps using CPU activation. BBs can be expanded in the field from 2 to 4 CPU sockets.
Each BB has up to 64 memory slots and can house up to 2TB memory, as well as 8 PCI Express Generation 3 slots.
Fujitsu building block technology provides step-by-step scalability up to 16 BBs or 1024 cores and 32TB of memory.
All Fujitsu M10 models have 8 slots for internal storage where hot swappable SAS hard drives and/or SSDs can
be installed. The onboard SAS controller supports RAID 0, 1 and 1E for hard disk drives.
4 Gigabit Ethernet ports, one 6Gbps SAS port and 2 USB ports are provided onboard.
Power supplies, Fans, and the unique Liquid Loop Cooling, or LLC for short, are configured with redundancy by default.
Note: Power supplies are 80 PLUS-compliant.
All Fujitsu M10 models support Solaris 11 and 10, as well as Solaris 9 and 8 in Solaris Legacy Containers.
All Fujitsu M10 models support Oracle VM Server for SPARC, or OVM, and Solaris Containers, also known as Zones.
In addition, Fujitsu M10-4S supports Physical Partitioning (PPAR for short) and PPAR dynamic reconfiguration.
In SPARC64 X (ten) and X+ (ten plus), multiple instruction combinations can be replaced by dedicated hardware, leading to much faster computing. This is called Software on Chip, or SWoC for short, and accelerates functions including:
Encryption/Decryption; where a dedicated cryptographic instruction executes encryption/decryption on the hardware at high-speed.
Decimal Floating Point Operations: Generally speaking computers perform arithmetic processing in binary numbers, whereas most of the numeric data handled by human beings are in decimal numbers. SPARC64 X (ten) and X+ (ten plus) processors can directly perform arithmetic processing of decimal numbers without the software having to perform the conversion to-and-from binary. This capability complies with industry standard IEEE754 (“I triple E seven fifty four”) and supports Oracle Number format adopted in Oracle Databases.
High Performance Computing Arithmetic Computational Extensions, HPC-ACE (“H – P – C - ace”) for short, and SIMD (“sim dee”):
HPC-ACE is an extension of the SPARC-V9 (“vee-nine”) architecture instruction set and was introduced in the SPARC64 VIII fx (sparc sixty-four-eight f-x) supercomputer processor at first.
HPC-ACE enhances the SIMD, Single Instruction Multiple Data, function.
SIMD is a technology that allows processing of multiple data in a single instruction. Multiple results for the price and time of one!
The Fujitsu M10-4S is a modular system that can be expanded by adding more Fujitsu M10-4S Building Blocks. The inter-processor interconnect reaches 25Gbps (giga bits per second). Overall, Fujitsu’s innovative interconnect technology from high end supercomputers ensures low latency and linear scalability up to 16 Building Blocks.
In Fujitsu M10-4S, the system bus between all Building Block chassis is implemented as a crossbar switch. To improve performance, the physical addressing of memory in the Building Block is evenly spread out across all CPUs in each Building Block. Processors have connections to the crossbar, and the crossbar supports data access and transfer to other Building Blocks.
Up to four Building Blocks can be directly connected using the crossbar built-in to each Fujitsu M10-4S chassis.
For larger configurations with system configurations scalable up to 16 Building Blocks, each chassis is connected to all other chassis via crossbar boxes.
The latest Fujitsu M10 servers are now equipped with the new SPARC64 X+ (ten plus) processor, providing users with up to 30% more processing power than the previous world record holder SPARC64 X (ten) processor. the new M10 server extends its lead further and now delivers approximately 2.8 times the CPU performance of IBM Power 795 servers.
The world record CPU performance is matched by the world record memory performance in the Fujitsu M10 servers. M10 delivers 4 TERABYTEs per sec memory bandwidth; nearly 5 times more memory performance compared to IBM’s best.
High speed memory and processing power, high capacity compute and memory capacity. Fujitsu M10 delivers the highest level of performance for resource intensive databases and enterprise applications.
Socket to socket, Fujitsu M10-4S delivers more performance and more value than IBM Power 795.
Fujitsu M10-4S with 16 sockets holds the world-record result on the two-tier SAP SD standard application benchmark running Oracle Database 11g Release 2 Enterprise Edition and outperforms IBM Power 795 at the same socket count.
The Fujitsu M10 can simplify server expansion by rapidly adding cpu resources using CPU Core Activation or COD (c-o-d).
With the CPU Activation feature, a system can be expanded in units of two cores with no system interruptions and no downtime.
The CPU Activation feature is available for all of Fujitsu M10 models.
Initial investments, for example, can be made to meet the current demand levels. Customers can then add CPU resources to their systems when needed in the future, ensuring business continuity with strong investment protection.
Core activation permits are transferable between separate, but same model, systems while they are both operating. You can do CPU activation transfers even between geographically separate sites. CPU Activations can be moved from the production system to the backup system during a disaster recovery scenario, for example.
As the heart of the server, the SPARC64 X (ten) and X+ (ten plus) processors contribute to the high reliability and the high uptime of Fujitsu M10 servers. With error detection and recovery functions embedded in all circuits, the processor can help continue server operations even after processor errors.
In the rare but much dreaded event of a complete CPU core failure, the Fujitsu M10 avoids performance degradation by automatically replacing the failed core with a reserve core, and by re-allocating the CPU Activation for the failed core to the reserved core. The result is improved uptime and service levels and continuous delivery of mission critical performance.
The Fujitsu M10-4S offers hardware partitioning for complete fault isolation. The granularity of these physical partitions, also called PPARS (“p-p-pars”) is a single Fujitsu M10-4S building block with 2 or 4 CPUs. There can be up to 16 building blocks in a Fujitsu M10-4S server so the maximum of physical partitions is 16. up to its maximum of 16 Building Blocks. If a database deployment is isolated within one hardware partition, a failure in any other partition will not affect the data base deployment. Essentially, what happens in a physical partition has no effect on any other physical partition. Note that physical partitions are available only on the Fujitsu M10-4S.
Oracle Solaris Predictive Self-Healing features on the host side and the Fujitsu predictive self-healing engine in the XSCF Service Processor work closely together to further enhance the reliability of the Fujitsu M10 systems. Among many RAS enhancements, Oracle Solaris Predictive Self-Healing software provides constant monitoring of CPUs and memory and off-lines components (cores, DIMM, CPU, …) that are experiencing soft errors before they actually fail.
In SPARC64 X (ten) and X+ (ten plus) almost all circuits are protected as indicated by the green in this diagram. Green means that detected errors are corrected, yellow means that errors are detected and reported so that data is not corrupted. Gray means that an error has no effect on the processor. Protected sections of the processor include – cache memories, arithmetical logic units, and registers. Protection includes both data error correction and instruction retry.
Improved RAS features include more than 10 times the number of error checkers, and ECC protected floating-point registers compared to the previous SPARC M8000/9000 series servers.
Even if an uncorrectable error is found, the OS and applications can continue to operate. The underlying concept is to keep as much of the resources functioning as possible. The minimum portion of the failing component is automatically degraded – for instance, a single core or a portion of cache memory can be marked inoperable while keeping the rest of the chip up and computing. Plus, recording processor events helps to detect any processor problems as early as possible.
All Fujitsu M10 servers include redundant and hot-swappable system components, diagnostic and error recovery features throughout the design, and built-in remote management features.
The advanced architecture of these reliable servers delivers high levels of application availability and rapid recovery from many types of hardware faults; simplifying system operation and lowering costs for enterprises.
Please refer to the RAS features shown on the slide. The blue squares are features designed to minimize unplanned downtime. The red circles are features that minimize planned downtime.
To highlight some of the most important RAS features that are unique to Fujitsu M10, please see the features in red.
Core Auto-Replacement allows failed cores to be automatically replaced from the reserves of unallocated cores, keeping the system running with no downtime or performance degradation.
Fujitsu M10 implements a memory redundancy mechanism, referred to as "Memory Mirroring“. This improves single system reliability by duplicating all data in memory, and protecting the system from uncorrectable multi-bit errors.
Hot swap and redundancy are implemented for fans, power, and disk storage systems.
The modular design and Physical Partition Dynamic Reconfiguration feature allow the Fujitsu M10-4S to grow by adding processors, memory and I/O without downtime.
Innovative Liquid Loop Cooling technology allows the system to be more compact with less heatsinks and fans, providing both higher performance as well as longer service life. And the pumps in the Liquid Loop Cooling system are redundantly configured.
XSCF (“x-s-c-f”), eXtended System Control Facility, provides an independent facility for hardware configuration, monitoring, and alerts. It offers several RAS enhancing features such as remote system control and console access over Ethernet.
And hardware RAID has also been added to all Fujitsu M10 models, giving customers choices when deciding how to protect data on onboard disk drives.
Oracle Enterprise Manager Ops Center is Oracle's single, integrated solution for managing all Oracle storage and sever assets in a data center.
Oracle Enterprise Manager Ops Center 12c provides asset discovery, asset provisioning, monitoring, patching,
and automated workflows. It can also discover and manage virtual servers as well as physical servers,
simplifying the management of high-end servers such as the Fujitsu M10-4S as well as all other SPARC
servers in a data center.
All Fujitsu M10 models are supported with Oracle Enterprise Manager Ops Center 12c Release 2 and higher.
Through its advanced server lifecycle management capabilities, Oracle Enterprise Manager Ops Center
12c provides a converged hardware management approach that integrates the management of servers,
storage, and network fabrics, including firmware, operating systems, and virtual machines.
The Fujitsu M10 offers virtualization and resource management technologies at no cost, built-in. This enables the deployment of many applications together to improve system utilization, optimize the use of computing resources, and deliver greater ROI from IT investments.
At the bottom of the virtualization stack are the SPARC64 X (ten) and X+ (ten plus) processors. The first level of virtualization, physical partitions or PPARs (“P pars”), are hardware partitions providing complete hardware fault and security isolation. PPARs (“P pars”) are available on the Fujitsu M10-4S model which scales from 1 to 16 Building Blocks. Each PPAR (“P par”) can have one or more Building Blocks making it a flexible infrastructure system with scalability up to 64 sockets, 1024 CPUs, and 32 TBs of memory!
In each PPAR (“P par”) are hypervisor-based Oracle VM (“v-m”) Server for SPARC partitions. They are also known as Logical Domains, or LDOMs (“L – doms”). Oracle VM Server for SPARC virtualization is available on all Fujitsu M10 models.
The next level of virtualization is Oracle Solaris Zones, a feature of Oracle Solaris. Each instance of Oracle Solaris has various resource management tools that are very useful for managing many applications on a single system image. With Oracle Solaris Zones there can be many partitions for each Solaris instance. This contrasts with Oracle VM Server for SPARC and PPARS (“p-pars”) where each LDOM (“L-dom”) or PPAR (“p-par”) has its own instance of Solaris. PPARS (“p-pars”) , LDOMs and Zones are designed to work together meaning that you can have Zones inside of LDOMs which can reside inside of PPARS (“p-par”) .
To highlight how that PPAR DR feature can be used here are some pictures.
Since December 2013, Fujitsu M10-4S has supported Physical Partition Dynamic Reconfiguration, PPAR (“p-par”) DR (“d-r”) for short. And this feature is supported now on the enhanced SPARC64 X+ (ten plus) systems as well.
PPAR (“p-par”) DR (“d-r”) enables customers to add a physical M10-4S chassis while the machine is running. As you know, SPARC Enterprise M9000 and M8000 also supported Dynamic Reconfiguration as well. With Dynamic Reconfiguration, we can grow a machine, hot swap building blocks in case of failure, and also rearrange physical partitions to meet the changing needs of the datacenter.
The value is flexibility. Grow, Repair, Adapt, all without disturbing production.
DR is available now. It is supported on all M10-4S servers, new ones and those already in the field. For M10 systems that were shipped before the end of 2013, DR just requires a firmware update and the latest OVM software version. Both are free.
PPAR (“p-par”) DR (“d-r”) and Oracle VM logical DR work together. As you know Oracle VM has been and continues to evolve. As Oracle VM DR features are enhanced, PPAR DR will also evolve. The Fujitsu and Oracle development teams are working closely together on these features.
Currently, Oracle VM DR does not allow for root complexes to be dynamically added or removed. Since PPAR (“p-par”) DR (“d-r”) works in conjunction with Oracle VM DR, the same restriction exists for PPAR DR. But, let’s not focus too much on this temporary limitation. Dynamic Reconfiguration is a great feature and will grow and expand just like M10-4S machines!
When onboard IO expansion is insufficient, Fujitsu M10 servers offer IO expansion with the PCI Expansion Unit.
This unit has 11 PCI Express Generation 3 slots, and each slot supports PCI Hot Plug.
All Fujitsu M10 models support connecting one or more PCI Expansion Units.
Fan units and power supplies in the PCI Expansion Unit are configured with redundancy by default.
The PCI Expansion Unit is connected to the M10 main unit through a link card.
The PCI Expansion Unit has proven to be a valuable feature for M10, and gives the M10 lineup a unique advantage when competing on IO slot count and for consolidation implementations.
The power-saving functions provided in Fujitsu M10 systems reduce the power being consumed by either setting a power cap or by reducing the power consumed by unused or low-utilization-rate hardware components.
The power capping function allows the customer to set an upper threshold for system power consumption. The CPU frequency is automatically controlled so as not to exceed the threshold. This function provides the customer with control of system power consumption to fit the datacenter facilities.
All Fujitsu M10 models also provide a Power Saving feature. Customers who prioritize lower power consumption can select the power-saving Elastic mode, whereas those who prefer performance can select the Performance mode. For M10-4S models, the power modes are set per physical partition.
Next, a comment about mixing old and new CPUs in M10 systems. This has been a frequently asked question.
There is no upgrade path from the SPARC64 X (ten) M10 systems to the SPARC64 X+ (ten plus) M10 systems.
For example, a SPARC64 X (ten) M10-1 cannot be upgraded to a SPARC64 X+ (ten plus) M10-1, and
a SPARC64 X (ten) M10-4S BB cannot be upgraded to a SPARC64 X+ (ten plus) M10-4S BB.
However, a SPARC64 X+ (ten plus) M10-4S BB can be connected to a SPARC64 X (ten) M10-4S BB
using an Expansion Kit or Expansion Rack.
The XSCF (“x-s-c-f”) in M10 will be familiar to SPARC Enterprise M-Series customers. Same easy interface. Same trusted quality.
The XSCF is a service processor that is independent of the Main Unit processors, memory and IO. The XSCF provides the following functions:
First: configuration management. The XSCF manages PPAR (“p-par”) configuration and CPU activation settings.
Second: power control of the main unit. The XSCF controls the powering on and off of the main unit.
If the XSCF detects an error in essential resources and decides that the main unit can't
continue to function properly, the XSCF prohibits the powering on of the main unit.
Next: Status monitoring. The XSCF monitors configuration and hardware abnormalities of the main unit
and various sensors throughout the machine.
And last: server power consumption monitoring. The XSCF monitors server power consumption
and provides power management functions in conjunction with OVM.
Here are some recent Fujitsu M10 updates to take note of.
Oracle Auto Service Request, or ASR (“a-s-r”), is now supported with Fujitsu M10 servers from firmware version XCP2050 (“x-c-p twenty fifty) or later and ASR release 4.6 or later.
Also, as mentioned earlier, Oracle Enterprise Manager Ops Center 12c is now supported on Fujitsu M10 servers from Release 2
And, PPAR DR (“p-par d-r”) is now supported on all M10-4S systems.
PPAR DR requires the following software versions, but no hardware changes are required.
Solaris 11.1 or later in the Control Domain.
SRU11.1.15 (“s-r-u eleven dot one dot fifteen”) or later is recommended.
System firmware version XCP2090 or later, and Oracle VM for SPARC (“sparc”) 3.0 or later.
PPAR DR is a no cost feature. All M10-4S systems in the field support PPAR DR, as long as the minimum software requirements are met.
In summary,
The Fujitsu M10 server is designed to take full advantage of the exceptional power and performance of the SPARC64 X (ten) and X+ (ten plus) processors.
Fujitsu’s building block technology provides step-by-step scalability up to 64 processors, 1024 cores and 32TB memory.
CPU Activation allows customers to add computing resources at the core level, as needed.
Mainframe class RAS is implemented across CPUs and the entire system in all Fujitsu M10 servers, including the M10-1.
All Fujitsu M10 servers support Solaris 11 and 10, as well as Solaris 9 and 8 in Solaris Legacy Containers.
Oracle VM for SPARC and Solaris Containers (also known as Zones) are supported by all Fujitsu M10 servers.In addition, M10-4S supports PPAR and PPAR dynamic reconfiguration.