IBM System z10 Business Class (z10 BC) Reference Guide The New Face of Enterprise Computing April 2009
Table of Contents IBM System z10 Business Class (z10 BC) Overview page 3 z/Architecture page 6 z10 BC page 11 z10 BC Design and Technology page 14 z10 BC Model page 15 z10 BC Performance page 17 z10 BC I/O Subsystem page 18 z10 BC Channels and I/O Connectivity page 19 HiperSockets page 34 Security page 36 Cryptography page 36 On Demand Capabilities page 41 Reliability, Availability, and Serviceability (RAS) page 45 Availability Functions page 46 Environmental Enhancements page
IBM System z10 Business Class (z10 BC) Overview In today’s world, IT is woven in to almost everything that Think Big, Virtually Limitless a business does and consequently is pivotal to a busi- The Information Technology industry has recognized the ness. Yet technology leaders are challenged to manage business value of exploiting virtualization technologies on sprawling, complex distributed infrastructures and the ever any and all server platforms.
More Solutions, More Affordable example, the zAAP and zIIP processors enable you to Today’s businesses with extensive investments in hardware purchase additional processing capacity exclusively for assets and core applications are demanding more from specific workloads, without affecting the MSU rating of the IT—more value, more transactions, more for the money. IBM System z model designation.
provide a more cost-effective, specialized z/OS® applica- The New Face Of System z tion Java execution environment. Workloads eligible for the IBM’s mainframe capabilities are legendary. Customers zAAP (with z/OS V1.8) include all Java processed via the deploy systems that remain available for years because IBM Solution Developers Kit (SDK) and XML processed they are expected to, and continue to, work above expec- locally via z/OS XML System Services. tations.
z/Architecture The z10 BC continues the line of upward compatible main- • Support for 1 MB page frames frame processors and retains application compatibility • Full hardware support for Hardware Decimal Floating- since 1964. The z10 BC supports all z/Architecture®-com- point Unit (HDFU) pliant Operating Systems. The heart of the processor unit is the IBM z10 Enterprise Quad Core processor chip run- z/Architecture operating system support ning at 3.
• Intelligent and optimized dispatching of workloads. With z/OS 1.9, IBM delivers functionality that continues to HiperDispatch can help provide increased scalability solidify System z leadership as the premier data server. and performance of higher n-way System z10 systems z/OS 1.9 offers enhancements in the areas of security, net- by improving the way workload is dispatched within the working, scalability, availability, application development, server.
z/VM • Improved availability with Parallel Sysplex and Coupling Facility improvement z/VM V5.
The z/VM hypervisor is designed to help clients extend the z/VSE 4.1 is designed to support: business value of mainframe technology across the enter- • z/Architecture mode only prise by integrating applications and data while providing • 64-bit real addressing and up to 8 GB of processor exceptional levels of availability, security, and operational storage ease.
enhanced access to open code libraries, both of which Operating System can be used to lower development costs. Large memory spaces can be used to increase both system and application efficiency as I/Os or memory management can be eliminated. z/OS V1R8, 9 and 10 ESA/390 (31-bit) z/Architecture (64-bit) No Yes No Yes Yes Yes No Yes No* Yes (1)(2) z/OS V1R7 with BM Lifecycle Extension for z/OS V1.
z10 BC The IBM System z10 Business Class (z10 BC) delivers higher availability and can be concurrently added or innovative technologies for small and medium enter- replaced when at least two drawers are installed. Reduced prises that give you a whole new world of capabilities capacity and priced I/O features will continue to be offered to run modern applications. Ideally suited in a Dynamic on the z10 BC to help lower your total cost of acquisition.
PUs defined as Internal Coupling Facilities (ICFs), Inte- be individually configured as a secure coprocessor or grated Facility for Linux (IFLs), System z10 Application an accelerator for SSL, the TKE workstation with optional Assist Processor (zAAPs) and System z10 Integrated Infor- Smart Card Reader, and provides the following CP Assist mation Processor (zIIPs) are no longer grouped together in for Cryptographic Function (CPACF): ™ ® one pool as on the IBM eServer zSeries 890 (z890), but • DES,
Additionally, several service enhancements have also mixed and unpredictable workload environments, provid- been designed to avoid unscheduled outages and include ing scalability, high availability and Qualities of Service continued focus on firmware quality, reduced chip count (QoS) to emerging applications such as WebSphere, Java on Single Chip Module (SCM) and memory subsystem and Linux. improvements.
z10 BC Design and Technology The System z10 BC is designed to provide balanced Speed and precision in numerical computing are important system performance. From processor storage to the for all our customers. The z10 BC offers improvements system’s I/O and network channels, end-to-end bandwidth for decimal floating point instructions, because each z10 is provided and designed to deliver data where and when processor chip has its own hardware decimal floating it is needed.
z10 BC Model Timed Interconnect Multiplexer (STI-MP) card. There are The z10 BC has one model, the E10, (Machine Type 2098) two types of HCA fanout cards: HCA2-C is copper and offering between 1 to 10 processor units (PUs), which is always used to connect to I/O (IFB-MP card) and the can be configured to provide a highly scalable solution HCA2-O which is optical and used for customer InfiniBand designed to meet the needs of both high transaction pro- coupling.
z10 BC model upgrades z10 BC Model Capacity IDs: The z10 BC provides for the dynamic and flexible capac- • A00, A01 to Z01, A01 to Z02, A03 to Z03, A04 to Z04, and A05 to Z05 ity growth for mainframe servers. There are full upgrades within the z10 BC and upgrades from any z9 BC or z890 to • Capacity setting A00 does not have any CP engines any z10 BC. Temporary capacity upgrades are available • Nxx, where n = the capacity setting of the engine, and through On/Off Capacity on Demand (CoD).
z10 BC Performance The performance design of the z/Architecture can enable • up to 1.4 times that of the z9 BC (2096) Z01. the server to support a new standard of performance for Moving from a System z9 partition to an equivalently sized applications through expanding upon a balanced system System z10 BC partition, a z/VM workload will experience approach.
z10 BC I/O Subsystem A new host bus interface using InfiniBand with a link data System I/O Configuration Analyzer rate of 6 GBps, was introduced on the z10 BC. It provides Today the information needed to manage a system’s I/O enough throughput to support the full capacity and pro- configuration has to be obtained from many separate cessing power of the CPC. The z10 BC contains an I/O applications.
z10 BC Channels and I/O Connectivity ESCON Channels FICON Express4 Channels The z10 BC supports up to 480 ESCON channels. The The z10 BC supports up to 128 FICON Express4 chan- high density ESCON feature has 16 ports, 15 of which nels, each one operating at 1, 2 or 4 Gb/sec auto-negoti- can be activated for customer use. One port is always ated. The FICON Express4 features are available in long reserved as a spare which is activated in the event of a wavelength (LX) and short wavelength (SX).
FICON Express Channels Concurrent Update The z10 BC also supports carrying forward FICON The FICON Express4 SX and LX features may be added Express LX and SX channels from z9 BC and z990 each to an existing z10 BC concurrently. This concurrent update channel operating at 1 or 2 Gb/sec auto-negotiated. Each capability allows you to continue to run workloads through FICON Express feature has two independent channels other channels while the new FICON Express4 features are (ports). being added.
The FICON Express4, FICON Express2 and FICON Express channels support native FICON and FICON Channel-to-Channel (CTC) traffic for attachment to servers, disks, tapes, and printers that comply with the FICON architecture. Native FICON is supported by all of the IBM z10 BC operating systems. Native FICON and FICON Two site non-cascaded director topology. Each CEC connects to directors in both sites. CTC are defined as CHPID type FC.
FCP – increased performance for small block sizes FICON and FCP for connectivity to disk, tape, and printers The Fibre Channel Protocol (FCP) Licensed Internal Code has been modified to help provide increased I/O High Performance FICON – improvement in performance and operations per second for small block sizes.
IBM System Storage DS8000 Release 4.1 delivers new Platform registration is a service defined in the Fibre Chan- capabilities to support High Performance FICON for nel – Generic Services 4 (FC-GS-4) standard (INCITS System z, which can improve FICON I/O throughput on a (ANSI) T11 group). DS8000 port by up to 100%. The DS8000 series Licensed Platform and name server registration applies to all of the Machine Code (LMC) level 5.4.2xx.
Extended distance FICON – improved performance at extended Exploitation of extended distance FICON is supported by distance IBM System Storage DS8000 series Licensed Machine An enhancement to the industry standard FICON architec- Code (LMC) level 5.3.1xx.xx (bundle version 63.1.xx.xx), ture (FC-SB-3) helps avoid degradation of performance at or later.
FICON Link Incident Reporting OSA-Express3 for reduced latency and improved throughput FICON Link Incident Reporting is designed to allow an To help reduce latency, the OSA-Express3 features now operating system image (without operating intervention) to have an Ethernet hardware data router; what was previ- register for link incident reports, which can improve the ously done in firmware (packet construction, inspection, ability to capture data for link error analysis.
The OSA-Express3 features are exclusive to System z10. There are operating system dependencies for exploitation of two ports in OSD mode per PCI-E adapter. Whether it is a 2-port or a 4-port feature, only one of the ports will be visible on a PCI-E adapter if operating system exploitation updates are not installed.
The OSA-Express3 and OSA-Express2 Ethernet features OSA-Express3 10 Gigabit Ethernet SR support the following CHPID types: The OSA-Express3 10 Gigabit Ethernet (GbE) short reach (LR) feature has two ports. Each port resides on a PCIe CHPID OSA-Express3, Type OSA-Express2 Features Purpose/Traffic OSC OSA-Integrated Console Controller (OSA-ICC) TN3270E, non-SNA DFT, IPL to CPC and LPARs Operating system console operations 1000BASE-T adapter and has its own channel path identifier (CHPID).
OSA-Express3-2P Gigabit Ethernet SX automatically adjusts to 10, 100, or 1000 Mbps, depending The OSA-Express3-2P Gigabit Ethernet (GbE) short upon the LAN. When the feature is set to autonegotiate, wavelength (SX) feature has two ports which reside on a the target device must also be set to autonegotiate. The single PCIe adapter and share one channel path identifier feature supports the following settings: 10 Mbps half or full (CHPID).
When configured at 1 Gbps, the 1000BASE-T Ethernet Internal “routing” can be disabled on a per QDIO connec- feature operates in full duplex mode only and supports tion basis. This support does not affect the ability to share jumbo frames when in QDIO mode (CHPID type OSD). an OSA-Express port.
OSA-Express Network Traffic Analyzer is exclusive to the OSA-Express2 (or OSA-Express3) port to the z/VM operat- z10 BC, z9 BC, z10 EC, and z9 EC, and is applicable ing system when the port is participating in an aggregated to the OSA-Express3 and OSA-Express2 features when group when configured in Layer 2 mode. Link aggregation configured as CHPID type OSD (QDIO), and is supported (trunking) is designed to allow you to combine multiple by z/OS.
Layer 2 transport mode: When would it be used? OSA Layer 3 Virtual MAC for z/OS If you have an environment with an abundance of Linux To simplify the infrastructure and to facilitate load balanc- images in a guest LAN environment, or you need to define ing when an LPAR is sharing the same OSA Media Access router guests to provide the connection between these Control (MAC) address with another LPAR, each operating guest LANs and the OSA-Express3 features, then using the system instance can now have
Direct Memory Access (DMA) CCL helps preserve mission critical SNA functions, such OSA-Express3 and the operating systems share a as SNI, and z/OS applications workloads which depend common storage area for memory-to-memory communi- upon these functions, allowing you to collapse SNA inside cation, reducing system overhead and improving perfor- a z10 BC while exploiting and leveraging IP. mance. There are no read or write channel programs for The OSA-Express3 and OSA-Express2 GbE and data exchange.
Token-Ring networks and ESCON channels with an Ether- Remove L2/L3 LPAR-to-LPAR Restriction net network and integrated LAN adapters on System z10, OSA port sharing between virtual switches can communi- OSA-Express3 or OSA-Express2 GbE or 1000BASE-T. cate whether the transport mode is the same (Layer 2 to OSA-Express for NCP is supported in the z/OS, z/VM, z/VSE, TPF, z/TPF, and Linux on System z environments. Layer 2) or different (Layer 2 to Layer 3).
HiperSockets The HiperSockets function, also known as internal Queued A HiperSockets device can filter inbound datagrams by Direct Input/Output (iDQIO) or internal QDIO, is an inte- Virtual Local Area Network identification (VLAN ID, IEEE grated function of the z10 BC server that provides users 802.1q), the Ethernet destination MAC address, or both.
interrupt. This enhancement is transparent to the operating HiperSockets Multiple Write Facility and zIIP enablement system in the receiving LPAR. HiperSockets Multiple Write is described as “zIIP-Assisted HiperSockets for large mes- Facility, with fewer I/O interrupts, is designed to reduce sages.” zIIP-Assisted HiperSockets can help make highly CPU utilization of the sending and receiving LPAR. secure, available, virtual HiperSockets networking a more attractive option.
Security Cryptography Today’s world mandates that your systems are secure and The z10 BC includes both standard cryptographic hard- available 24/7. The z10 BC employs some of the most ware and optional cryptographic features for flexibility and advanced security technologies in the industry—helping growth capability.
Enhancements to CP Assist for Cryptographic Function (CPACF): Crypto Express2 Accelerator – for Secure Sockets Layer (SSL) acceleration: CPACF has been enhanced to include support of the fol• Is designed to support clear-key RSA operations lowing on CPs and IFLs: • Offloads compute-intensive RSA public-key and private• Advanced Encryption Standard (AES) for 192-bit keys key cryptographic operations employed in the SSL pro- and 256-bit keys tocol Crypto Express2 features can be carried forward • SHA-
Cryptographic enhancements to Crypto Express2 and Crypto Express2-1P Improved key exchange – Added Improved key exchange with non-CCA cryptographic systems. New features added to IBM Common Cryptographic Architecture (CCA) are designed to enhance the ability to exchange Dynamically add crypto to a logical partition.
Support for AES encryption algorithm includes the master Enhancement with TKE 5.3 LIC key management functions required to load or generate The TKE 5.3 level of LIC includes support for the AES AES master keys, update those keys, and re-encipher key encryption algorithm, adds 256-bit master keys, and tokens under a new master key.
TKE additional smart cards – new feature Remote Key Loading Benefits You have the capability to order Java-based blank smart • Provides a mechanism to load initial ATM keys without the need to send technical staff to ATMs cards which offers a highly efficient cryptographic and data management application built-in to read-only memory • Reduces downtime due to key entry errors for storage of keys, certificates, passwords, applications, • Reduces service call and key management costs and data.
On Demand Capabilities It may sound revolutionary, but it’s really quite simple. In The new contract set is structured in a modular, hierarchi- the highly unpredictable world of On Demand business, cal approach. This new approach will eliminate redundant you should get what you need, when you need it. And you terms between contract documents, simplifying the con- should pay for only what you use. Radical? Not to IBM. It’s tracts for our customers and IBM.
can be added up to what the physical hardware model While all new CBU contract documents contain the new can contain for up to 10 days for a test activation or 90 CBU Test terms, existing CBU customers will need to exe- days for a true disaster recovery. cute a contract to expand their authorization for CBU Test upgrades if they want to have the right to execute produc- On system z10 the CBU entitlement records contain an tion workload on the CBU Upgrade during a CBU Test.
• The quantity of temporary CP capacity ordered is limited customer, via the Resource Link ordering process, deter- by the quantity of purchased CP capacity (permanently mines how many tokens go into each pool. Once On/Off active plus unassigned) CoD resources are activated, tokens will be decremented • The quantity of temporary IFLs ordered is limited by from their pools every 24 hours.
z/OS Capacity provisioning allows you to set up rules Capacity on Demand – Permanent Capacity defining the circumstances under which additional capac- Customer Initiated Upgrade (CIU) facility: When your ity should be provisioned in order to fulfill a specific busi- business needs additional capacity quickly, Customer ness need. The rules are based on criteria, such as: a Initiated Upgrade (CIU) is designed to deliver it.
Reliability, Availability, and Serviceability (RAS) In today’s on demand environment, downtime is not only with the introduction of concurrent I/O drawer add and unwelcome—it’s costly. If your applications aren’t consis- eliminating pre-planning requirements. These features are tently available, your business suffers.
Availability Functions Redundant I/O Interconnect With the z10 BC, significant steps have been taken in the area of server availability with a focus on reducing pre- In the event of a failure or customer initiated action such planning requirements. Pre-planning requirements are as the replacement of an HCA/STI fanout card, the z10 BC minimized by delivering and reserving 8 GB for HSA so the is designed to provide access to your I/O devices through maximum configuration capabilities can be exploited.
Dynamic Oscillator Switchover fied in the configuration tool along with a “starting” logical The z10 BC has two oscillator cards, a primary and a memory size. The configuration tool will then calculate the backup. For most cases, should a failure occur on the pri- physical memory required to satisfy this target memory.
Environmental Enhancements Power and cooling discussions have entered the budget IBM Systems Director Active Energy Manager planning of every IT environment. As energy prices have IBM Systems Director Active Energy Manager™ (AEM) is a risen and utilities have restricted the amount of power building block which enables customers to manage actual usage, it is important to review the role of the server in bal- power consumption and resulting thermal loads IBM serv- ancing IT spending.
Parallel Sysplex Cluster Technology IBM System z servers stand alone against competition and Although there is significant value in a single footprint and have stood the test of time with our business resiliency multi-footprint environment with resource sharing, those solutions. Our coupling solutions with Parallel Sysplex customers looking for high availability must move on to technology allow for greater scalability and availability. a database data sharing configuration.
16 allows one of these protocol exchanges to complete Coupling Facility Configuration Alternatives asynchronously. This allows faster duplexed request ser- IBM offers multiple options for configuring a functioning vice time, with more benefits when the Coupling Facilities Coupling Facility: are further apart, such as in a multi-site Parallel Sysplex • Standalone Coupling Facility: The standalone CF environment.
Introducing long reach InfiniBand coupling links Now, InfiniBand can be used for Parallel Sysplex coupling and STP communication at unrepeated distances up to 10 z/OS ICF ICF km (6.2 miles) and even greater distances when attached z/OS System z10 / z9 zSeries 990 / 890 to a qualified optical networking solution.
The IBM System z10 BC also takes advantage of System z now supports 12x InfiniBand single data rate InfiniBand as a higher-bandwidth replacement for the Self- (12x IB-SDR) coupling link attachment between System Timed Interconnect (STI) I/O interface features found in z10 and System z9 general purpose (no longer limited to prior System z servers. standalone coupling facility) 5) Long Reach 1x InfiniBand coupling links (1x IB- InfiniBand coupling links are CHPID type CIB.
z10 Coupling Link Options Type Description Use The Sysplex Timer Model 2 is the centralized time source Link data rate PSIFB 1x IB-DDR LR z10 to z10 5 Gbps PSIFB 12x IB-DDR IC Internal Coupling Channel z10 to z10 z10 to z9 Distance ICB-4 Copper connection between OS and CF z10, z9 2 GBps z990, z890 ISC-3 Fiber connection between OS and CF z10, z9 2 Gbps z990, z890 that sets the Time-Of-Day (TOD) clocks in all attached z10 Max servers to maintain synchronization.
join the two server CTN. Previously, if both the Preferred STP External Time Source (ETS), the time of an STP-only Time Server (PTS) and the Backup Time Server (BTS) Coordinated Timing Network (CTN) can track to the time experienced a simultaneous power outage (site failure), provided by the NTP server, and maintain a time accuracy or both experienced a POR, reinitialization of time, and of 100 milliseconds. special roles (PTS, BTS, and CTS) was required.
Continuous Availability of NTP servers used as Exter- attaching NTP servers to the SE LAN. The HMC, via a nal Time Source: Improved External Time Source (ETS) separate LAN connection, can access an NTP server avail- availability can now be provided if you configure different able either on the intranet or Internet for its time source. NTP servers for the Preferred Time Server (PTS) and the Note that when using the HMC as the NTP server, there is Backup Time Server (BTS).
Application Programming Interface (API) to automate Two data centers STP CTN reconfiguration: The concept of “a pair and • CTN with 2 servers (one in each data center) install IBF a spare” has been around since the original Sysplex on at least the PTS/CTS Couple Data Sets (CDSs). If the primary CDS becomes – Also recommend IBF on BTS to provide recovery protection when BTS is the CTS unavailable, the backup CDS would take over.
HMC System Support The new functions available on the Hardware Management HMC/SE support is addressing the following requirements: Console (HMC) version 2.10.1 as described apply exclu- • The availability of addresses in the IPv4 address space sively to System z10. However, the HMC version 2.10.1 will is becoming increasingly scarce. continue to support the systems as shown. • The demand for IPv6 support is high in Asia/Pacific The 2.10.
There is now a new Console Manager task that offers Enhanced installation support for z/VM using the HMC basic messaging capabilities to allow system operators or HMC version 2.10.1 along with Support Element (SE) administrators to coordinate their activities. The new task version 2.10.1 on z10 BC and corresponding z/VM 5.4 sup- may be invoked directly, or via a new option in Users and port, will now give you the ability to install Linux on System Tasks.
Implementation Services for Parallel Sysplex IBM Implementation Services for Parallel Sysplex CICS and This DB2 data sharing service is designed for clients who WAS Enablement want to: IBM Implementation Services for Parallel Sysplex Middle- 1) Enhance the availability of data ware – CICS enablement consists of five fixed-price and 2) Enable applications to take full utilization of all servers’ fixed-scope selectable modules: resources 1) CICS application review 3) Share application system resour
Fiber Quick Connect for FICON LX Environments GDPS Fiber Quick Connect (FQC), an optional feature on z10 Geographically Dispersed Parallel Sysplex (GDPS) is BC, is offered for all FICON LX (single-mode fiber) chan- designed to provide a comprehensive end-to-end con- nels, in addition to the current support for ESCON (62.5 tinuous availability and/or disaster recovery solution micron multimode fiber) channels.
z10 BC Physical Characteristics Physical Planning z10 BC System Power A System z10 BC feature may be ordered to allow use of the z10 BC in a non-raised floor environment. This capability may help ease the cost of entry into the z10 BC; a raised 1 I/O Drawer 2 I/O Drawers 3 I/O Drawers 4 I/O Drawers normal room (<28 degC) 3.686 kW 4.542 kW 5.308 kW 6.253 kW floor may not be necessary for some infrastructures. warm room (>=28 degC) 4.339 kW 5.315 kW 6.291 kW 7.
z10 BC Configuration Detail Features Min # Features Max # Features Max Connections Increments Purchase per Feature Increments 16-port ESCON 0 (1) 32 480 channels 16 channels 4 channels 1reserved as as a spare FICON Express4* 0 (1) 32 64/128* channels 2/4* channels 2/4* channels FICON Express2** 0 (1) 20 80 channels 4 channels 4 channels FICON Express** 0 (1) 20 40 channels 2 channels 2 channels ICB-4 0 (1) 6 12 links (2) (3) 2 links 1 link ISC-3 0 (1) 4 links 1 link 1
z10 BC Model Structure z10 BC System weight and IBF hold-up times z10 Model E10 – Single Frame Model PU PUs for Max Avail Standard Standard CP/IFL/ Max Max Customer Subcapacity SAPs Spares ICF/zAAP/ Customer Chan. CPs zIIP** Memory E10 4 10 5 2 0 5/10/10/5/5 248 GB 480* * Max is for ESCON channels. ** For each zAAP and/or zIIP installed there must be a corresponding CP. The CP may satisfy the requirement for both the zAAP and/or zIIP.
Coupling Facility – CF Level of Support CF Level Function z10 EC z10 BC z9 EC z9 BC z990 z890 16 CF Duplexing Enhancements List Notification Improvements Structure Size increment increase from 512 MB –> 1 MB X 15 Increasing the allowable tasks in the CF from 48 to 112 14 CFCC Dispatcher Enhancements X X 13 DB2 Castout Performance X X 12 z990 Compatibility 64-bit CFCC Addressability Message Time Ordering DB2 Performance SM Duplexing Support for zSeries X X X X X X 11 z990 Compatibilit
Statement of Direction IBM intends to support optional water cooling on future The System z10 will be the last server to support connec- high end System z servers. This cooling technology will tions to the Sysplex Timer (9037). Servers that require time tap into building chilled water that already exists within the synchronization, such as to support a base or Parallel Sys- datacenter for computer room air conditioning systems. plex, will require Server Time Protocol (STP).
Publications The following Redbook publications are available now: z10 BC Technical Overview Hardware Management Console Operations Guide (V2.10.
© Copyright IBM Corporation 2009 IBM Systems and Technology Group Route 100 Somers, NY 10589 U.S.A Produced in the United States of America, 04-09 All Rights Reserved References in this publication to IBM products or services do not imply that IBM intends to make them available in every country in which IBM operates. Consult your local IBM business contact for information on the products, features, and services available in your area.
