Capacity Management (ITILv3)

Capacity Management (ITILv3) Marek Rychlý Vysoké uˇcení technické v Brneˇ Fakulta informaˇcních technologií Ústav informaˇcních systému˚

Pˇrednáška pro ISE 30. bˇrezna 2015

Marek Rychlý

Capacity Management (ITILv3) — Pˇrednáška pro ISE, 30. bˇrezna 2015

1 / 34

Obsah

1

Opakování a cíle pˇrednášky Capacity (& Performance) Management Cíle pˇrednášky

2

Capacity (& Performance) Management Capacity and Performance Management Aktivity a komunikace Spolupráce služeb

Marek Rychlý


2 / 34

Opakování a cíle pˇrednášky Capacity (& Performance) Management

Capacity (& Performance) Management Cíle pˇrednášky

Capacity (& Performance) Management Obsah: Zajištuje nákladoveˇ optimální kapacitu služeb odpovídající souˇcasným i budoucím potˇrebám. ˇ Reší otázky kapacity a výkonu služeb i jejich zdroju. ˚ ˇ (výšší efektivita výdaju, ˚ redukce rizik, duv ˚ eryhodnost investiˇcního plánování)

Vede Capacity Mgmt. IS a s jeho pomocí spravuje Capacity Plan. ..................................................................... v ESM patˇrí C&P Mgmt. mezi služeby IT Infrastructure Mgmt. (služba pro údržbu a ˇrízení IT infrastruktury)

v ITILv2 patˇrí Capacity Mgmt. mezi procesy Service Delivery (procesy zajišt’ující dostupnost a kvalitu IT služeb)

v ITILv3 patˇrí Capacity Mgmt. mezi procesy Service Design ˇ a vylepšení stávajících IT služeb) (procesy pro návrh nových nebo zmeny

Marek Rychlý


4 / 34


Capacity (& Performance) Management Cíle pˇrednášky

Cíle pˇrednášky

Znát cíle, obsah a pˇrínosy Capacity (& Performance) Mgmt. ˇ popsat role, aktivity a komunikaci v C&P Managementu. Umet Znát metriky v této oblasti vˇc. tzv. „Key Performance Indicators“. ˇ (napˇr. uvést pˇríklad nekolika takových metrik)

Mít pˇredstavu o spolupráci služeb v této oblasti.

Marek Rychlý


5 / 34


Capacity and Performance Management Aktivity a komunikace Spolupráce služeb

Capacity (& Performance) Management

zajišt’uje, že je a vždy bude k dispozici odpovídající IT kapacita (ˇreší jen dostupnost kapacity; dostupnost samotné služby ˇreší Availability Mgmt.)

výkon služeb musí odpovídat nebo být vyšší než požadují SLAs (monitoruje a spravuje kapacitu a výkon služeb a jejich zdroju) ˚

asistuje pˇri diagnóze a vyˇrešení problému˚ s výkonem a kapacitou ˇ kapacity sestavuje Capacity Plan, ˇreší dopad pˇrípadných zmen realizuje pˇrípadná vylepšení služeb a vyhodnocuje jejich pˇrínos

Marek Rychlý


7 / 34



Požadavky a omezení ˇ Kvalitu služeb mohou ovlivnovat „nefunkˇcní“ požadavky zákazníka, (rychlost, dostupnost, udržovatelnost, kontrolovatelnost, pˇrenositelnost, bezpeˇcnost, lokalizace apod.)

omezení, která vycházejí z „funkˇcních“ požadavku˚ zákazníka. (geografická, technické nebo legislativní standardy, existující IT prostˇredí apod.)

ˇ ritelné! Nefunkˇcní požadavky musí být oveˇ požadavky jsou souˇcástí Service Level Agreement (SLA), použití vhodných metrik kvality poskytovaných služeb, (napˇr. „v 90 % musí být do 1 hodiny možnost zarezervovat 1 TB místa v úložišti“)

ˇ neplést s metrikami uspešnosti C&P Managementu. (tzv. „Key Performance Indicators“, bude popsáno dále)

Marek Rychlý


8 / 34



Požadavky vs. možnosti

(diagramy pˇrevzaty z „Capacity and Performance Management Primer“)

Marek Rychlý


9 / 34



Metriky

ˇ a poˇcet I/O operací na stroji, využití CPU, obsazení pametí I/O sít’ový provoz, statistika pruchozích ˚ paketu˚ a spojení, využití sít’ové kapacity na jednotlivých rozhraních a uzlech, stav front na routerech, zahazování paketu˚ na firewallech, poˇcet uživateli provedených akcí za jednotku cˇ asu, (de facto vytížení informaˇcního systému uživateli)

ˇ jednotlivých doba odezvy systému na akci uživatele a pomer cˇ ástí systému˚ na zpracování odezvy, (de facto odezva informaˇcního systému a jeho cˇ ástí)

???

Marek Rychlý


10 / 34



Key Performance Indicators (KPIs, dle ITILu) ˇ Metriky uspešnosti C&P Managementu: Incidents due to Capacity Shortages (poˇcet incidentu˚ v dusledku ˚ nedostateˇcné kapacity služby nebo IT komponenty)

Exactness of Capacity Forecast (odchylky naplánované od skuteˇcneˇ spotˇrebovabné kapacity, chyby plánování)

Capacity Adjustments (poˇcet úprav služeb/komponent s cílem plánovaneˇ pˇrizpusobit ˚ jejich kapacitu)

Unplanned Capacity Adjustments (poˇcet úprav služeb/komponent pro ˇrešení neoˇcekávaných problému˚ s kapacitou)

Resolution Time of Capacity Shortage (doba ˇrešení neoˇcekávaných problému˚ s kapacitou)

Capacity Reserves ˇ eˇ a maximálneˇ spotˇrebovávané kapacity oproti celkové dostupné) (podíly bežn

Percentage of Capacity Monitoring ˇ sledovaných služeb/komponent oproti všem službám/komponentám) (pomer Marek Rychlý


11 / 34



Dílˇcí procesy v rámci C&P Managementu 1

Business Capacity Management (pˇrevádí business potˇreby a plány na budoucí kapacitní a výkonostní požadavky pro služby a IT komponenty)

2

Service Capacity Management (monitoruje, ˇrídí a pˇredvídá kapacitu a výkon služeb, provádí operativní a ˇ služeb dle potˇreby; napˇr. ˇreší neoˇcekávané situace špiˇcky) preventivní zmeny

3

Component Capacity Management (monitoruje, ˇrídí a pˇredvídá kapacitu, výkon a rozsah využití IT komponent a ˇ dle potˇreby; napˇr. ladí poˇradí zdroju, ˚ provádí jejich operativní a preventivní zmeny ˇ eˇ vytížená) a priority úkolu˚ tak, aby byla IT infrastruktura soustavneˇ prum ˚ ern

4

Capacity Management Reporting (zprostˇredkovává informace ostatním službám, vytváˇrí tzv. „Capacity Reports“)

Marek Rychlý


12 / 34



Uspoˇrádání podprocesu˚ C&P Managementu

(diagram pˇrevzat z „Insight: Is Performance Engineering Primarily a Business Issue?“)

Marek Rychlý


13 / 34



Prostˇredky C&P Managementu Capacity Management Information System ˇ (virtuální úložišteˇ dat, pˇredpovedí, plánu˚ kapacity a reportu; ˚ nad ruznými ˚ dbs.)

Capacity Plan (plán zdroju˚ pro poskytnutí IT služeb v požadovaném rozsahu, obsahuje scénaˇre ˇ pro ruzné pro ruzné ˚ pˇredpovedi ˚ business požadavky)

Capacity Report (poskytuje informace ostatním službám, pˇrehled a plán vytížení služeb a zdroju, ˚ ˇ pˇrehled úspešnosti pˇredchozího plánování vzhledem ke skuteˇcnostem)

Capacity Forecast (pˇredvídá budoucí požadavky na kapacitu a výkon služeb/komponent, trendy)

Suggestion to Influence Service Demand (návrhy na zvyšování kapacity a výkonu služeb, projednávají se výše v rámci vyšování kvality služeb; napˇr. nahrazení hardware, reorganizace procesu, ˚ atd.)

Marek Rychlý


14 / 34



Databáze kapacity

(diagram pˇrevzat z „ITRO Disciplines: Capacity Management“)

Marek Rychlý


15 / 34



Role a týmy v C&P Managementu Capacity Manager (zaruˇcuje nákladoveˇ optimální a vhodneˇ naˇcasované poskytování služeb v dohodnuté kapaciteˇ a výkonu, spravuje zdroje služeb, schvaluje krátkodobé, ˇ stˇrednedobé a dlouhodobé plány)

„PerfCap“ týmy (capacity manager, správci a vlastníci zdroju, ˚ zástupci zákazníka)

ˇreší požadavky na zmeny ˇ rezervované kapacity a rychlosti, historické události, neoˇcekávané špiˇcky a jejich ˇrešení, ˇ konfigurace pro lepší C&P Management, navrhují zmeny (tj. „Suggestion to Influence Service Demand“)

ˇ a využití kapacity a rychlosti, krátkodobeˇ plánují zajištení (plány cˇ inností a vytížení zdroju, ˚ postupy pro operativní ˇrízení)

ˇ a využití kapacity a rychlosti. dlouhodobeˇ plánují zajištení ˇ ení ˇ dlouhodobých trendu) (plány pro usmern ˚

Marek Rychlý


16 / 34



Aktivity v rámci C&P Managementu 1

analýza stavající kapacity a výkonu služby, (sledování metrik z SLAs, incidentu˚ a problému, ˚ atd.)

2

ˇ návrh na zmenu/vylepšení služby nebo komponenty, ˇ z business plánu, (identifikace požadavku, ˚ napˇr. diskuzí s uživateli, pˇredpovedí ˚ trendu˚ využítí, požadavku˚ na nové služby, atd.)

3

ohodnocení, odsouhlasení a dokumentace nových požadavku, ˚ (porada s PerCap týmy a jejich doporuˇcení, odsouhlasení Capacity Managerem)

4

zanesení do plánu kapacity, (vˇc. využití potˇrebných zdroju, ˚ vyhodnocení dopadu, ˚ popisu možných scénáˇru) ˚

5

ˇ sledování úspešnosti, pˇrípadneˇ revize plánu. (vyhodnocení KPIs, synchronizace s aktuálními business požadavky a plány)

Marek Rychlý


17 / 34


Capacity and Performance Management Aktivity a komunikace

Spolupráce The interpretation process can be represented asslužeb a flowchart, as shown in Figure 1-1. You can set this practice up all by yourself or automate it by using PM for System i or PM for System p.

Pˇríklad: Performance Management Workflow

Set the base line

Collect data

Produce reports

Analyze reports

CPU

Memory

Disk

Network

Jobs / Processes

Compare with baseline

(diagram pˇrevzat z „A Systems Management Guide to Performance Management for System i and System p servers“, IBM)

Figure 1-1 A systems management view of performance management

Marek Rychlý


18 / 34



ˇ Cinnosti v rámci C&P Managementu (dle ITILv2) I

Workload and Performance Management (monitorování kapacity a výkonu IT komponent, operativní ˇrízení dle požadavku˚ ˇ zákazníka, neoˇcekávaných situací, behem špiˇcek a dle detekovaných trendu) ˚

Application Sizing (nastavení požadavku˚ aplikace na základeˇ dostupných informací a historie, urˇcení podmínek pro detekci rizika nedostateˇcné kapacity a rychlosti)

Resource Management ˇ poˇradí a priorit jednotlivých aktivit tak, aby byla IT (evidence zdroju, ˚ ladení ˇ eˇ vytížená) infrastruktura soustavneˇ prum ˚ ern

Marek Rychlý


19 / 34



ˇ Cinnosti v rámci C&P Managementu (dle ITILv2) II

Demand Management (zpracování požadavku˚ na aktuální i budoucí cˇ innosti)

Modeling (analýza vzájemných vlivu˚ požadavku˚ na IT zdroje, jejich modelování a ˇ organizace za úˇcelem schválení požadavku, ˚ optimalizace provádení)

Capacity Planning (plánování využití kapacity, evidence plánu˚ a skuteˇcností, vykrytí denních a týdenních špiˇcek)

Marek Rychlý


20 / 34



Demand Management ˇ analyzuje, pˇredvídá a ovlivnuje požadavky zákazníku˚ na služby, (tak, aby kapacita služeb vždy pokryla požadavky)

samostatný proces v rámci ITIL 2011. (ale velmi úzce spolupracuje s Capacity Mgmt.)

(diagram pˇrevzat z „Activity-based Demand Management“)

Marek Rychlý


21 / 34



Infrastructure Service Interconnections (ESM) Configuration Management ˇ (pˇri plánování se využívá aktuální konfigurace a navrhují její zmeny)

Event Management ˇ na riziko nedostateˇcné kapacity nebo výkonu) (upozornení

Operations Management (operativní ˇrízení kapacity a výkonu)

Availability Management (dostupnost služby je nutná, teprve pak zaleží na její kapaciteˇ a výkonu)

Inventory (evidence IT komponent, jejich vlastností)

Business Process Management ˇ dostupné kapacity a výkonu na business aktivity zákazníka) (vliv zmen

Resource Management (evidence zdroju˚ dostupných pro plánování kapacity a výkonu) Marek Rychlý


22 / 34



Relationship Service Interconnections (ESM) Reporting Management ˇ požadavku˚ je hlášeno) (vytížení IT komponent a splnení

SLA Management (požadavky na kapacitu a výkon jsou zakotveny v SLA)

Knowledge Management (zaznamenává se vývoj dostupné kapacity a výkonu, závislosti na cˇ innostech)

Asset Management (kapacita a výkon muže ˚ ovlivnit náklady nebo napˇr. licence)

Notification and Escalation Management (problémy jsou pˇredávány k ˇrešení jednotlivým SME)

Problem Management (problémy s dostupnou kapacitou a výkonem je tˇreba ˇrešit)

Change Management ˇ kapacity a výkonu) (nutnost zmeny Marek Rychlý


23 / 34



Ukázkový scénáˇr spolupráce služeb 1

ˇ Operátor zaznamená, že server behem poslední hodiny vykazuje vysoké vytížení CPU. (Event Management, Operations Management, Problem Management)

2

ˇ Správce systému potvrdí, že bylo tˇreba server vymenit, protože je už nevyhovující. (Inventory, Configuration Management)

3

Analyzují se business aktivity, zda-li muže ˚ být jejich nedávná ˇ zmena pˇríˇcinou souˇcasného problému. (Business Process Management, Resource Management)

4

ˇ zátež ˇ na mezi stávající Výsledkem analýzy je doporuˇcení rozdelit a další (nový) server. (Resource Management, Reporting Management, Change Management)

5

ˇ vytížení CPU puvodního Poté, co byl nový server zprovoznen, ˚ ˇ na pˇrijatelnou hodnotu. serveru se snížilo zpet (Configuration Management, Knowledge Management)

Marek Rychlý


24 / 34



Proces DS3 Manage Performance and Capacity v COBIT 4.1 — zde má být pˇríloha — pˇrejít na pˇrílohu

Marek Rychlý


25 / 34



Sít’ová kapacita a výkon

ˇ 80 % nedostateˇcné rychlosti Více než a dostupnosti aplikace bývá pˇrisuzováno sít’ovým problémum. ˚

Ale sít’ové problémy jsou pravou pˇríˇcinou jen v méneˇ než 20 % pˇrípadu. ˚

Marek Rychlý


26 / 34



Nástroje pro P&C Management

Cacti (SNMP orientated performance management tool with RRDTool-based graphs)

BMC Patrol and Perform & Predict HP’s Openview Microsoft Operations Manager Computer Associate’s Unicenter Suite (performance monitoring a automatizace v IBM z/OS systems management)

Marek Rychlý


27 / 34



Detailed Report Graph for IBM i — zde má být pˇríloha — pˇrejít na pˇrílohu

Další pˇríklady nástroju˚ od IBM: IBM Tivoli Omegamon (formerly Candle for mainframe) (performance monitoring pro širokou škálu aplikací)

IBM TotalStorage Productivity Center Suite (TPC, Capacity) (hromadneˇ spravuje využití diskových úložišt’ a databází, vˇc. dostupnosti)

IBM Tivoli Monitoring (ITM, Performance) (detekuje „bottlenecks“ a problémy v systémových zdrojích, ˇreší kritické situace)

IBM z/OS Workload Manager/System Resources Manager (WLM/SRM)

Marek Rychlý


28 / 34

ˇ Shrnutí a záver ˇ Podekování a otázky Pˇrílohy

ˇ Shrnutí a záver

C&P Mgmt. ˇreší nákladoveˇ optimální kapacitu služeb odpovídající souˇcasným i budoucím potˇrebám. ˇ Potˇrebná kapacita a výkon jsou dány vnejšími požadavky, je nutno použít vhodné metriky. Cílem je efektivní využívání IT komponent a neustálé zlepšování kvality poskytovaných služeb.

Marek Rychlý


29 / 34


ˇ Dekuji za pozornost.

Otázky? Diskuze?

Marek Rychlý


30 / 34


COBIT 4.1: Process DS3 Manage Performance and Capacity ˇ ˇ — ZACÁTEK PRÍLOHY — ˇ do prezentace zpet

pˇreskoˇcit pˇrílohu

Proces v doméneˇ „Deliver and Support“. ˇ Jeden z nejduležit ˚ ejších procesu. ˚ Velice podobný (a kompatibilní) s ITIL procesem Capacity Management. Pˇrevzato z „IT Governance Institute: COBIT 4.1. ISACA, 2007, ISBN 1-933284-72-2“.

Marek Rychlý


31 / 34

Deliver and Support Manage Performance and Capacity

DS3

PROCESS DESCRIPTION DS3 Manage Performance and Capacity

Eff ec tiv Eff ene s i Co cien s nfi cy de Int ntial eg ity Av rity ail Co abili mp ty l Re ianc lia e bil ity

The need to manage performance and capacity of IT resources requires a process to periodically review current performance and capacity of IT resources. This process includes forecasting future needs based on workload, storage and contingency requirements. This process provides assurance that information resources supporting business requirements are continually available.

P

P

S

Control over the IT process of Manage performance and capacity that satisfies the business requirement for IT of optimising the performance of IT infrastructure, resources and capabilities in response to business needs by focusing on meeting response time requirements of SLAs, minimising downtime, and making continuous IT performance and capacity improvements through monitoring and measurement is achieved by • Planning and providing system capacity and availability • Monitoring and reporting system performance • Modelling and forecasting system performance and is measured by • Number of hours lost per user per month due to insufficient capacity planning • Percent of peaks where target utilisation is exceeded • Percent of response-time SLAs not met

CE MANENT FOR PER SUREM MEA

IT GOVERNANCE

RESOURCE MANAGEMENT

Primary

Ap pli ca Inf tion orm s Inf atio ras n t Pe ructu op le re

DE VAL LIV UE ER Y

MAN RISK AGE MEN T

C GI T TE EN RA M ST IGN AL

Secondary

© 2007 IT Governance Institute. All rights reserved. www.itgi.org

109

DS3


CONTROL OBJECTIVES DS3 Manage Performance and Capacity DS3.1 Performance and Capacity Planning Establish a planning process for the review of performance and capacity of IT resources to ensure that cost-justifiable capacity and performance are available to process the agreed-upon workloads as determined by the SLAs. Capacity and performance plans should leverage appropriate modelling techniques to produce a model of the current and forecasted performance, capacity and throughput of the IT resources. DS3.2 Current Performance and Capacity Assess current performance and capacity of IT resources to determine if sufficient capacity and performance exist to deliver against agreed-upon service levels. DS3.3 Future Performance and Capacity Conduct performance and capacity forecasting of IT resources at regular intervals to minimise the risk of service disruptions due to insufficient capacity or performance degradation, and identify excess capacity for possible redeployment. Identify workload trends and determine forecasts to be input to performance and capacity plans. DS3.4 IT Resources Availability Provide the required capacity and performance, taking into account aspects such as normal workloads, contingencies, storage requirements and IT resource life cycles. Provisions such as prioritising tasks, fault-tolerance mechanisms and resource allocation practices should be made. Management should ensure that contingency plans properly address availability, capacity and performance of individual IT resources. DS3.5 Monitoring and Reporting Continuously monitor the performance and capacity of IT resources. Data gathered should serve two purposes: • To maintain and tune current performance within IT and address such issues as resilience, contingency, current and projected workloads, storage plans, and resource acquisition • To report delivered service availability to the business, as required by the SLAs Accompany all exception reports with recommendations for corrective action.

110



DS3

MANAGEMENT GUIDELINES DS3 Manage Performance and Capacity From Inputs

Outputs

Availability, continuity and recovery specification System monitoring requirements SLAs

AI3 DS1

To

Performance and capacity information PO2 PO3 Performance and capacity plan (requirements) PO5 AI1 Required changes AI6 Process performance reports ME1

Functions

Bus

Establish a planning process for the review of performance and capacity of IT resources. Review current IT resources’ performance and capacity. Conduct IT resources’ performance and capacity forecasting. Conduct gap analysis to identify IT resources mismatches. Conduct contingency planning for potential IT resources unavailability. Continuously monitor and report the availability, performance and capacity of IT resources.

CIO

Bus in

Activities

CFO

CEO

ess E

xec u

tive

RACI Chart

AI3 ME1

ines s Pr oce Hea ss O dO wne per r atio ns Chi ef A rch itec t Hea dD eve lopm Hea ent d IT Adm inis PM trat O ion Com p Ris lian k a ce, nd Au Sec dit, urit y

AI2

A C C C C

I C I I

R A/R A/R A/R A/R

I

I

A/R

C C

C C C R C

C C C C C

C C C C I

I C

I

I

I

I

A RACI chart identifies who is Responsible, Accountable, Consulted and/or Informed.

Goals and Metrics

Goals

IT

Process

• Respond to business requirements in alignment with the business strategy. • Make sure that IT services are available as required. • Optimise the IT infrastructure, resources and capabilities.

set

Metrics

measure

• Peak load and overall utilisation rates • Percent of peaks where target utilisation is exceeded • Percent of response-time SLAs not met • Transaction failure rate


e driv

• Number of hours lost per user per month due to insufficient capacity planning • Number of critical business processes not covered by a defined service availability plan

• Planning and providing system capacity and availability • Monitoring and reporting system performance • Modelling and forecasting system performance

set

e driv

measure

Activities

• Monitor and measure peak load and transaction response times. • Meet response-time SLAs. • Minimise transaction failures. • Minimise downtime. • Optimise utilisation of IT resources.

measure

• Frequency of performance and capacity forecasting • Percent of assets included in capacity reviews • Percent of assets monitored through centralised tool(s)

111

DS3


MATURITY MODEL DS3 Manage Performance and Capacity Management of the process of Manage performance and capacity that satisfies the business requirement for IT of optimising the performance of IT infrastructure, resources and capabilities in response to business needs is: 0 Non-existent when Management does not recognise that key business processes may require high levels of performance from IT or that the overall business need for IT services may exceed capacity. There is no capacity planning process in place. 1 Initial/Ad Hoc when Users devise workarounds for performance and capacity constraints. There is very little appreciation of the need for capacity and performance planning by the owners of the business processes. Action taken toward managing performance and capacity is typically reactive. The process for planning capacity and performance is informal. The understanding of current and future capacity and performance of IT resources is limited. 2 Repeatable but Intuitive when Business and IT management are aware of the impact of not managing performance and capacity. Performance needs are generally met based on assessments of individual systems and the knowledge of support and project teams. Some individual tools may be used to diagnose performance and capacity problems, but the consistency of results is dependent on the expertise of key individuals. There is no overall assessment of the IT performance capability or consideration of peak and worst-case loading situations. Availability problems are likely to occur in an unexpected and random fashion and take considerable time to diagnose and correct. Any performance measurement is based primarily on IT needs and not on customer needs. 3 Defined when Performance and capacity requirements are defined throughout the system life cycle. There are defined service level requirements and metrics that can be used to measure operational performance. Future performance and capacity requirements are modelled following a defined process. Reports are produced giving performance statistics. Performance- and capacity-related problems are still likely to occur and be time-consuming to correct. Despite published service levels, users and customers may feel sceptical about the service capability. 4 Managed and Measurable when Processes and tools are available to measure system usage, performance and capacity, and results are compared to defined goals. Up-to-date information is available, giving standardised performance statistics and alerting incidents caused by insufficient performance and capacity. Insufficient performance and capacity issues are dealt with according to defined and standardised procedures. Automated tools are used to monitor specific resources, such as disk space, networks, servers and network gateways. Performance and capacity statistics are reported in business process terms, so users and customers understand IT service levels. Users feel generally satisfied with the current service capability and may demand new and improved availability levels. Metrics for measuring IT performance and capacity are agreed upon but may be only sporadically and inconsistently applied. 5 Optimised when The performance and capacity plans are fully synchronised with the business demand forecasts. The IT infrastructure and business demand are subject to regular reviews to ensure that optimum capacity is achieved at the lowest possible cost. Tools for monitoring critical IT resources are standardised and used across platforms and linked to an organisationwide incident management system. Monitoring tools detect and can automatically correct performance- and capacity-related issues. Trend analysis is performed and shows imminent performance problems caused by increased business volumes, enabling planning and avoidance of unexpected issues. Metrics for measuring IT performance and capacity have been fine-tuned into outcome measures and performance indicators for all critical business processes and are consistently measured. Management adjusts the planning for performance and capacity following analysis of these measures.

112



COBIT 4.1: Process DS3 Manage Performance and Capacity ˇ — KONEC PRÍLOHY — ˇ na zaˇcátek pˇrílohy zpet

Marek Rychlý


32 / 34


Detailed Report Graph1 for IBM i ˇ ˇ — ZACÁTEK PRÍLOHY — ˇ do prezentace zpet

pˇreskoˇcit pˇrílohu

Executive & Management Summary, Response Time (incl. Local Terminals & PC), Transaction Volumes & Analysis, Jobs, Print Outs, Processor Utilization & Trend, Memory & Disk Space Usage (incl. Disk Arm Utilization), Communication Lines, Additional Facts. 1 pˇrevzato Marek Rychlý

z http://www-03.ibm.com/systems/power/support/pm/ Capacity Management (ITILv3) — Pˇrednáška pro ISE, 30. bˇrezna 2015

33 / 34

Executive Summary ABC Corporation December 2003 - February 2004 Time frame 08.00 - 17.30

SWEDEN (10-ABCDE) 9406 820-2395/1521 CPU(s) 1 Internal memory in MB 1024 Disk Capacity in GB 67

Processor - Interactive Capacity Critical Marginal Acceptable

02/04

03/04

05/04

08/04

03/04

05/04

08/04

03/04

05/04

08/04

03/04

05/04

08/04

Processor - System+Interactive Critical Marginal Acceptable

02/04

Processor - Total Critical Marginal Acceptable

02/04

Disk Space Critical Marginal Acceptable

02/04

Printed 03/2004

Report 150

Management Summary SWEDEN (10-ABCDE) 9406 820-2395/1521 CPU(s) 1 Internal memory in MB 1024 Disk Capacity in GB 67

ABC Corporation December 2003 - February 2004 Time frame 08.00 - 17.30

Processor - Interactive Capacity Avg. 50.49 Peak 91.54 02/04

04/04

06/04

Acceptable

08/04

10/04

Marginal

12/04

02/05

Critical

Processor - System+Interactive Avg. 9.88 Peak 17.61 02/04

04/04

06/04

Acceptable

08/04

10/04

Marginal

12/04

02/05

Critical

Processor - Total Avg. 47.60 Peak 69.19 02/04

04/04

06/04

Acceptable

08/04

10/04

Marginal

12/04

02/05

Critical

Disk Space Avg. 73.83

02/04

04/04 Acceptable

Printed 03/2004

06/04

08/04 Marginal

10/04

12/04

02/05

Critical

Report 175

Response Time ABC Corporation December 2003 - February 2004 Time frame 08.00 - 17.30

SWEDEN (10-ABCDE) 9406 820-2395/1521 CPU(s) 1 Internal memory in MB 1024 Disk Capacity in GB 67

Average response time in seconds

Average Response Time per Measured Day 100

10

1

0.1

0.01 12/01

12/11

12/21

12/31 01/10 01/20 01/30 Date (all measured days are displayed)

02/09

02/19

02/29


Average Response Time by the Hour 1

0.1

0.01 06

07

08

09

10

11

12

13 14 15 16 17 18 19 20 21 22 The starting time for each hour is indicated

23

00

01

02

03

04

05

Facts Month December -03 January -04 February -04

Average Response Time 0.3 0.3 0.3

See Fact Sheet

Printed 03/2004

Report 250

Response Time, Local Terminals SWEDEN (10-ABCDE) 9406 820-2395/1521 CPU(s) 1 Internal memory in MB 1024 Disk Capacity in GB 67



Average Response Time per Measured Day Local Terminals 100

10

1

0.1

0.01 12/01

12/11

12/21


02/09

02/19

02/29


Average Response Time by the Hour Local Terminals 1

0.1 06

07

08

09

10

11

12


23

00

01

02

03

04

05


Printed 03/2004


Report 260

Response Time, PC SWEDEN (10-ABCDE) 9406 820-2395/1521 CPU(s) 1 Internal memory in MB 1024 Disk Capacity in GB 67



Average Response Time per Measured Day PC 0.1

0.01 12/01

12/11

12/21


02/09

02/19

02/29


Average Response Time by the Hour PC 0.1

0.01 06

07

08

09

10

11

12


23

00

01

02

03

04

05


Printed 03/2004


Report 280

Transaction Volumes SWEDEN (10-ABCDE) 9406 820-2395/1521 CPU(s) 1 Internal memory in MB 1024 Disk Capacity in GB 67


Transaction Volume per Hour

Thousands

Transactions per hour

15

10

5

0 12/01

12/11

12/21

12/31

01/10

01/20 Date

The average hour

01/30

02/09

02/19

02/29

The hour with the most transactions

Transaction Volume per Hour - History and 3 Month Trend 4 Thousands

Transactions per hour

5

3 2 1 0 02/03

04/03

06/03

08/03

10/03

12/03

Average number of transactions per hour

02/04

04/04

Trend

Facts Last 12 months 16.0

Average Transaction Growth in % Last 6 months Last 3 months 19.0 21.9

Last month 20.1

See Fact Sheet

Printed 03/2004

Report 350

Transaction Analysis SWEDEN (10-ABCDE) 9406 820-2395/1521 CPU(s) 1 Internal memory in MB 1024 Disk Capacity in GB 67


Transactions per Hour All Users 8

Thousands

6

4

2

0 06

07

08

09

10

11

12


23

00

01

02

03

04

05

All Users 1000 800 600 400 200 0

1 0.1 0.01 06

07

08

09

10

11

12

13 14 15 16 17 18 19 20 21 22 23 00 01 02 03 The starting time for each hour is indicated Total number of I/Os per transaction CPU time per transaction in seconds Number of synchronous I/Os per transaction

04

05

CPU-time per trans. in seconds

Number of I/O's per transaction

I/O and CPU Usage for the Average Transaction


The period's average transaction contains CPU-time in seconds Number of I/Os Number of synch. I/Os 0.20 104 46 0.03 25 17 0.04 32 22

See Fact Sheet

Printed 03/2004

Report 375

Jobs SWEDEN (10-ABCDE) 9406 820-2395/1521 CPU(s) 1 Internal memory in MB 1024 Disk Capacity in GB 67


Number of Jobs per Measured Day

Number of jobs

1000

100

10 12/01

12/11

12/21

12/31 Interactive jobs

01/10

01/20 Date

01/30

02/09

02/19

02/29

Batch jobs

600

400

200

0

0.003 0.002 0.001 0.000 06

07

08

09

10

11

12

13 14 15 16 17 18 19 20 21 22 23 00 01 The starting time for each hour is indicated Total number of I/Os per second CPU usage in seconds per I/O Number of synchronous I/Os per second

02

03

04

05


Batch jobs during the period containing CPU-time per I/O Number of I/Os per sec Number of synch. I/Os per sec 0.0008 416 212 0.0010 401 202 0.0008 405 212

See Fact Sheet

Printed 03/2004

Report 450

CPU-time per I/O in seconds

Number of I/O's per second

I/O and CPU Usage for Batch Jobs

Print Outs SWEDEN (10-ABCDE) 9406 820-2395/1521 CPU(s) 1 Internal memory in MB 1024 Disk Capacity in GB 67


Average number of pages per hour

Daily Average per Hour 1000

100

10

1 12/01

12/11

12/21

12/31

01/10

01/20 01/30 02/09 Date Total number of pages written to out queues per hour Number of interactive pages written to out queues per hour

02/19

02/29

Facts Print jobs with the most pages to out queues during the period

Job PRTJOB001 PRTJOB002 PRTJOB003 PRTJOB004 PRTJOB005 PRTJOB006 PRTJOB007 PRTJOB008 PRTJOB009 PRTJOB010

Number of pages 45445 7096 3451 3220 2308 2303 1896 1414 1392 1267

User with the most print outs to out queues during the period

User PRTUSR001 PRTUSR002 PRTUSR003 PRTUSR004 PRTUSR005 PRTUSR006 PRTUSR007 PRTUSR008 PRTUSR009 PRTUSR010

Number of pages 112689 5019 4987 4887 3644 3298 2822 2724 2533 2093

Printed 03/2004

Report 550

Processor Utilization SWEDEN (10-ABCDE) 9406 820-2395/1521 CPU(s) 1 Internal memory in MB 1024 Disk Capacity in GB 67


296

80

222

60

148

40

74

20

0 12/01

Usage in % (average)

CPW

Average Processor Utilization in % per Day

0 12/11

12/21

12/31

Interactive

01/10

System

01/20 Date Batch

01/30

02/09

02/19

02/29

Int. Peak Avg

370

100

296

80

222

60

148

40

74

20

0

0 06

07

08

09

Interactive

10

11

12

13

System

14

15

16

17

18

Batch

19

20

21

22

23

Int. Peak Max Interactive Limit

Facts Recommended guideline in %, interactive priority Recommended guideline in %, system + interactive priority Recommended guideline in %, all priorities

Printed 03/2004

00

01

02

03

04

05

Tot. Peak Max

Avg. Marginal 40 45 80

Critical 55 60 90

Report 650


CPW

Usage By the Hour, Average During the Period

Processor Utilization, Trend SWEDEN (10-ABCDE) 9406 820-2395/1521 CPU(s) 1 Internal memory in MB 1024 Disk Capacity in GB 67


Peak Processor Utilization, Total 370

100

296

80

222

60

148

40

74

20

0


CPW

Three month trend, the two heaviest hours per measured day

0 02/03

04/03 Acceptable Usage(average)

06/03 08/03 Marginal

10/03 Critical

12/03

02/04

04/04 Trend Trend(3,6,12)

Peak Interactive CPU 370

100

296

80

222

60

148

40

74

20

0

0 02/03

04/03 Acceptable Usage(average)

06/03 08/03 Marginal

10/03 12/03 Critical Interactive Limit

02/04


Facts Recommended guideline in %, interactive priority Recommended guideline in %, system+int. priority Recommended guideline in %, all priorities Number of months until the guideline is reached, interactive priority Number of months until the guideline is reached, system + int. priority Number of months until the guideline is reached, all priorities

Printed 03/2004

Peak 48 56 85 12 12 4

62 70 94 12 12 6

Report 675


CPW

Three month trend, the two heaviest hours per measured day

Memory SWEDEN (10-ABCDE) 9406 820-2395/1521 CPU(s) 1 Internal memory in MB 1024 Disk Capacity in GB 67


Percent of Time Faulting, Per Day 40

Percent

30

20

10

0 12/01

12/11

12/21

12/31

Average

01/10

01/20 Date

01/30

02/09

02/19

02/29

Peak

Percent of Time Faulting, Per Hour

Percent

30

20

10

0 06

07

08

09

10

11

Average

12

13 14 15 16 17 18 19 20 21 22 The starting time for each hour is indicated Peak

23

00

01

02

03

04

05

Guideline

Facts Recommended Guideline in %

Printed 03/2004

Marginal 20

Critical 25

Report 750

Disk Space Usage SWEDEN (10-ABCDE) 9406 820-2395/1521 CPU(s) 1 Internal memory in MB 1024 Disk Capacity in GB 67


Disk Space Usage in % per Day Usage in % (average)

100 80 60 40 20 0 12/01

12/11

12/21

12/31

01/10

01/20 Date

01/30

02/09

02/19

02/29

Disk Space Usage in %, 3 Month Trend Usage in % (average)

100 80 60 40 20 0 02/03

04/03 06/03 08/03 Acceptable Marginal Usage in % (average)

10/03 Critical

12/03

02/04


Facts Recommended guideline in % Number of months until the guideline is reached Change of stored data in % during the period Maximum usage of disk space in % during the period

Printed 03/2004

70 0

80 12 -5.81 88.18

Report 850

Disk Arm Utilization SWEDEN (10-ABCDE) 9406 820-2395/1521 CPU(s) 1 Internal memory in MB 1024 Disk Capacity in GB 67


Peak Disk Arm Utilization in % The two heaviest hours per measured day


20

15

10

5

0 12/01

12/11

12/21

12/31

01/10

01/20 Date

01/30

02/09

02/19

02/29

Peak Disk Arm Utilization in %, 3 Month Trend Usage in % (average)

100

10

1

0.1 02/03

04/03 06/03 08/03 Acceptable Marginal Usage in % (average)

10/03 Critical

12/03

02/04


Facts Recommended guideline in % Number of months until the guideline is reached

Printed 03/2004

Peak 30

40 12

Report 875

Communication Lines SWEDEN (10-ABCDE) 9406 820-2395/1521 CPU(s) 1 Internal memory in MB 1024 Disk Capacity in GB 67


Line Utilization COMLINK01 COMLINK01 COMLINK02 COMLINK02 CONLINK01 CONLINK02 ROCKLIN1 ROCKLIN1 ROCKLIN2 ROCKLIN2 SWITCHLIN1 SWITCHLIN1 0

10

20

December -03 Error

Printed 03/2004

30

40 50 60 Usage in % (average) January -04

70

80

90

100

February -04

Report 950

Additional Facts Report 250 - Response Time

Report 250 - Response Time

Users with the longest response times

Users with the longest average response time

User

Seconds

User

VSAMMONS RJADRAN QSECOFR QSECOFR RBARTLEY RBARTLEY PWERTZLER KTHOMPSON GREG PWERTZLER

8858.00 4684.00 3752.00 1499.00 700.00 407.00 287.00 206.00 196.50 153.00

RJADRAN RJADRAN RBARTLEY GREG RJADRAN RJADRAN RJADRAN RBARTLEY RJADRAN RBARTLEY

Seconds 66.92 26.46 6.84 4.02 1.31 0.84 0.81 0.67 0.58 0.56

Report 350 - Transaction Volumes

Report 375 - Transaction Analysis

Users with the most transactions during the period

Users with the most I/Os per transactions

User

User

Transactions

VSAMMONS KTHOMPSON GREG PWERTZLER RJADRAN ADMIN1 ADMIN2 VSAMMONS VSAMMONS ADMIN3

66826 59530 57572 47332 41670 40266 38736 38322 35880 31095

Number of I/Os

RJADRAN GREG PWERTZLER KTHOMPSON ADMIN02 USR01 VSAMMONS ADMIN03 USR010 ADMIN04

2239 1730 95 94 91 91 80 80 79 74

Report 375 - Transaction Analysis

Report 450 - Batch jobs

Users with the most CPU-time per transaction

Jobs with the most CPU-time during the period

User

Job

Seconds

VSAMMONS RJADRAN GREG KTHOMPSON ADMIN1 ADMIN2 ADMIN3 RJADRAN USR01 USR01

1.51 0.80 0.32 0.31 0.31 0.18 0.12 0.11 0.10 0.07

USRJOB01 USRJOB05 USRJOB06 USRJOB50 USRJOB51 USRJOB52 USRJOB07 USRJOB53 USRJOB08 USRJOB12

Seconds 38368 26995 14507 8306 5720 4615 4423 3521 2849 2439



Jobs with the most I/Os per second

The most frequent jobs

Job

Job

Number of I/Os

USRJOB10 USRJOB14 ADMJOB01 USRJOB12 ADMJOB02 ADMJOB12 USRJOB05 USRJOB13 USRJOB34 USRJOB35

Printed 03/2004

732 621 365 347 332 313 311 301 276 270

USRJOB54 USRJOB55 USRJOB56 USRJOB57 USRJOB58 USRJOB59 USRJOB60 USRJOB61 USRJOB62 USRJOB35

Number of starts 1857 1091 1091 108 93 65 58 58 54 54

Fact Sheet 1

ABC Corporation Technical Specifications

SWEDEN (10-ABCDE)

Model Internal memory in MB Disk Capacity in GB

9406 820-2395/1521 CPU(s) 1 1024 67

Period December 2003 - February 2004 Business Days M Tu W Th F Time frame 08.00 - 17.30 Changes in the model during the period Changes in the internal memory during the period Changes in the disk capacity during the period

No No No

Missing Report Days 02/05

Trends Number of months used for statistics

Management Summary CPU Int. Capacity

CPU Sys+Int

CPU Total

Disk DB CPU

Memory

Avg. Peak Max Avg. Peak Max Avg. Peak Max Avg. Max Avg. Peak Max Avg. Peak Max

12

Utilization in % 50.49 91.54 226.50 9.88 17.61 51.80 47.60 69.19 82.15 73.83 88.18 22.03 34.79 65.34 8.15 15.26 25.34

Months 12 12 12 12 12 6 12 0 0 0 0 0

System Values QACTJOB QADLACTJ QTOTJOB QADLTOTJ QPFRADJ Avg.Active Jobs Max.Active Jobs

Printed 03/2004

100 10 50 10 2 710 1653

Fact Sheet 2


Detailed Report Graph for IBM i ˇ — KONEC PRÍLOHY — ˇ na zaˇcátek pˇrílohy zpet

Marek Rychlý


34 / 34

Capacity Management (ITILv3)

Recommend Documents