KIT WS10-11 Introduction
Total Page:16
File Type:pdf, Size:1020Kb
High Availability and Scalability with System z and z/OS Joachim von Buttlar, Robert Vaupel IBM Deutschland Research & Development GmbH © 2010 IBM Corporation Who is Who? Joachim von Buttlar – System z Firmware Development – [email protected] Robert Vaupel – z/OS Workload Management Development and Design – IBM Senior Technical Staff Member – [email protected] 2 © 2010 IBM Corporation WS 2010/2011: Structure and Content CPU Architecture z/OS – Register sets – Memory organization – Address space concept – Virtual storage – Task execution and serialization – Interrupt mechanism – Program communication and data exchange – Timing facilities – Data formats, data sets and I/O flow – Instruction set – Multiprocessing facilities – z/OS subsystems: TSO, ISPF, JES I/O Architecture z/OS Dispatching and Hiperdispatch – I/O infrastructure – Adapter types & channels z/OS Workload Management – Control unit & devices – Extensions for large configurations Parallel Sysplex Partitioning and virtualization – Cluster concepts – LPAR versus z/VM – Parallel Sysplex structure and exploitation – Differences and commonalities – Data Mirroring and Global Dispersed Parallel Sysplex – Hardware facilities – Data mirroring – Storage management – Processor management – I/O management Middleware Integration and Software Architecture Date (always Fridays) 11:30-13:00 14:00-15:30 22.10.2010 Introduction and Orientation System z Architecture 5.11.2010 System z Architecture System z Architecture 19.11.2010 System z Architecture System z Architecture 3.12.2010 z/OS Introduction z/OS Introduction 17.12.2010 z/OS Dispatching and Virtualization z/OS Dispatching and Virtualization 14.1.2011 z/OS Parallel Sysplex z/OS Workload Management 28.1.2011 z/OS Workload Management z/OS and System z Software Architecture 11.2.2011 Wrap-Up and Closing 3 © 2010 IBM Corporation What is System z? IBM z Enterprise z196 IBM System z10 EC System /360 AA System System z z server server is is what what businessesbusinesses use use to to host host the the largest largest commercialcommercial databases, databases, transaction transaction servers,servers, and and applications applications that that requirerequire a a greater greater degree degree of of security security andand availability availability than than is is commonly commonly foundfound on on smaller-scale smaller-scale machines. machines. 4 © 2010 IBM Corporation System z Architecture S/360 architecture is based on von S/360 = 360° Neumann‘s computing model: One hardware architecture One operating system For all IBM computers 360° 315° 45° S/360 architecture got invented and documented in the S/360 Principles of Operation in 1964 by: 270° 90° – Gene Amdahl – Fred Brooks – Garry Blaauw 225° 135° http://publibz.boulder.ibm.com/epubs/pdf/dz9zr007.pdf 180° 5 © 2010 IBM Corporation System z and z/OS History MVT SVS MVS/370 MVS/XA MVS/ESA OS/390 z/OS MFT Expanded Parallel Java Storage Sysplex Websphere Virtual I/O Workload IEEE Float Fast Management 64 bit Fixed Storage One Address Spaces 2 GB Program Unix System IRD 15 Partitions 16MB Virtual Load Services Hiperdispatch VS Multiple Virtual Storage Dynamic I/O TCP/IP Offload or Tasks Area Storage Posix ... Security Cluster GDPS ... 1960 1965 1970 1975 1980 1985 1990 1995 2000 2005 Expanded CMOS 64bit Storage Access Technology Addressing 2GB Addressing Registers Parallel LPAR Data-spaces Sysplex Symmetric Multi Processing 7. April 1964 Virtual Memory Introduction of S/360 Architecture S/360 S/370 S/390 z Architecture 6 © 2010 IBM Corporation Mainframe Computing Mainframes are computers which – Execute hundreds of applications – Connect to thousands of I/O devices – And serve thousands of users simultaneously Mainframes can best be defined by their characteristics – The most important characteristic is to ensure a reliable and predictable execution of transactions – The importance of mainframes is for data base transaction processing and as the backend in data centers 7 © 2010 IBM Corporation Economical Importance: Why System z and z/OS All companies which have the need to store huge amounts of data require – Security – Scalability – Compatibility – Availability – Reliability – Serviceability 95% of the 2000 world-wide biggest companies use System z computers Around 65-70% of all relevant data are stored on System z computers 60% of all data being access thru the world wide web are stored in databases on System z (DB2, VSAM, and IMS) 8 © 2010 IBM Corporation High Availability and Scalability System z Hardware Overview and Introduction System z Usage in Customer Environments RAS capabilities What does High Availability and Scalability mean? 9 © 2010 IBM Corporation Elements of System z Architecture Central Processor Units – Up to 64 PUs Main Memory – Byte-wise addressable – 64-bit addressability – 'shared' between all CPU‘s I/O Subsystem – ‘old': parallel (copper), 4.5 MB/sec – 1990: serial (fiber), 17 MB/sec – 1999: FiCON (fiber), 270 MB/sec ESCON & FiCON 'Director' – Switch Control Units (CU) – Managing unit Devices – Hard disk, tape, printer, etc. Network (GbE, ...) 10 © 2010 IBM Corporation Heart of System z Architecture: MCM MCM = Multi Chip Module – Processor Units (PU), Storage Controller (SC), SEEPROM (S) and clock functions – Integration increases with each generation, example: • z9: 8 PUs per MCM with up to 2 cores • z10: 5 PUs per MCM with up to 4 cores • z196: 6 Pus per MCM with up to 4 cores A single MCM can provide 24 processors on a z196 but a z196 can have up to 96 processors (80 usable for workloads) 11 © 2010 IBM Corporation z196 PU chip, SC chip and MCM 12 © 2010 IBM Corporation z196 Book Layout MCM @ 1800W 8 I/O FAN OUT Backup Air Plenum 16X DIMMs Refrigeration Cooled or 2 FSP 100mm High Water Cooled M MC Fanout Memory Front Rear Cards Memory DCA Power Supplies 3x DCA 11 VTM Card Assemblies 14X DIMMs 8 Vertical 100mm High 3 Horizontal Cooling from/to MRU 13 © 2010 IBM Corporation z196 Water cooled Under the covers (Model M66 or M80) front view 14 © 2010 IBM Corporation z196 Frames On z196: Traditional System z Operating Systems: z/OS, Linux, zVSE, zVM On z196 Blade Extensions: Power 7 Blades, System x Blades Integration via Unified Resource Manager 15 © 2010 IBM Corporation Growth of System z Servers Growth encompasses – Speed: from z900 (770MHz) to z196 (5.2 GHz) – Integration of processors and chips on same MCM – Number of MCMs per system – And now with z196 • Integration of Blade Server z/OS release used for LSPR measurements z196 measurements are for a xx-way PCI - (Processor Capacity Index 16 © 2010 IBM Corporation A typical System z could look like this L D J LL L L L L L V z/ C I S ii i i i Ci Ci z/ z/ i L S O C B I M A B a n n n M M V V n n n n n n n E S I a C S P uu u u u Su Su i S S u C t 2 v S x x x n E E x S c x x x x x x h a u x z/VM z/VSE z/OS z/VMz/VM V4 LPAR LPAR LPARLPAR LPAR LPAR CP1 CP2 CP3 CP4 zIIP zAAP IFL1 IFL2 IFL3 Standard Processors Offload Engines Linux Engines System z Enterprise Server 17 © 2010 IBM Corporation System z Processor Characterization Central Processor ( CP ) – Provides processing capacity for z/Architecture and ESA/390 instruction sets – Runs z/OS, z/VM, z/VSE, z/TPF, Linux for System z System Assist Processor ( SAP ) – SAPs manage the start and ending of I/O operations for all LPARs and all attached I/O – Each machine has at least one SAP Internal Coupling Facility ( ICF , since 1997) – Provides additional processing capacity for the execution of the Coupling Facility Control Code (CFCC) in a CF LPAR Integrated Facility for Linux ( IFL , since 2001) – Provides additional processing capacity for Linux workloads IBM System z Application Assist Processors ( zAAP , since 2004) – Provides additional processing capacity for Java workloads under z/OS IBM System z Information Integration Processors ( zIIP , since 2006) – Provides additional processing capacity for certain DB2 workloads under z/OS Spares – Provides extra processing capacity in case of any failure of any PU SAP, ICF, IFL, zAAP, zIIP offer the same functionality as CPs Lower price than CP Do not affect traditional System z software charges 18 © 2010 IBM Corporation Why is System z different? Many different types of workloads Business Critical workloads Running systems at very high utilizations Access to systems is always required 19 © 2010 IBM Corporation System z Quality of Services RAS – Reliability – Availability – Serviceability Security / Integrity Scalability Manageability – Centralized control – Workload management Virtualization / Partitioning Technology – Workload separation Capacity – Evolving architecture Flexibility / Variety – Multiple workloads, multiple users Compatibility Capability – Autonomic features 20 © 2010 IBM Corporation System z: RAS Design Focus High Availability (HA) – The attribute of a system designed to provide service during defined periods, at acceptable or agreed upon levels and masks UNPLANNED OUTAGES from end-users . It employs fault tolerance, automated failure detection, recovery, bypass reconfiguration, testing, problem and change management . Continuous Operations (CO) – Attribute of a system designed to continuously operate and mask PLANNED OUTAGES from end-users. It employs non-disruptive hardware and software changes, non-disruptive configuration, software coexistence. Continuous Availability (CA) – Attribute of a system designed to deliver non-disruptive service to the end user 7 days a week, 24 HOURS A DAY (there are no planned or unplanned outages). It includes the ability to recover from a site disaster by switching computing to a second site. High Continuous Continuous Availability Availability Operations 21 © 2010 IBM Corporation Business Issue of “Non-Availability“ On demand challenges – Downtime unaffordable – Heterogeneous by nature – Complex to manage Loss of business Loss of customers – the Unplanned Outage Causes Application Failures competition is just a mouse click Hardw are 30% 45% away Failures 25% Loss of credibility, brand image IDC 2005 and stock value Operator Errors E.g.