GS Gruppen 7. juni 2011 High Availability, Disaster Recovery & More… For Windows, Linux, AIX & IBM i Danni Stjernegaard Director Nordic +45 502 06 danni. stjernegaard@visionsolutions. com Leaders Have Vision™ visionsolutions. com 1
Kort introduktion - Startede hos Vision Solutions 1. september 2010, som nordisk ansvarlig. - Tidligere job - Salgsdirektør hos ATEA - Direktør for Informatikk Denmark - Salgschef hos Scribona - IT-specialist Techline AS/400 IBM Denmark Leaders Have Vision™ visionsolutions. com 2
Hvem er Vision Solutions? • Markedslederende leverandør af højtilgængelig og system management software – – Windows, Linux, Power Systems og Cloud Computing Strategisk partnersamarbejde med IBM, Microsoft, HP, Dell, VMware m. fl. • Startede med udvikling af high availability, disaster recovery, og data management software for IBM Power Systems • Stort udviklings- og support team • – – 24/7 support “follow the sun” 350+ medarbejdere Worldwide partner netværk – – 800+ partnere, 25, 000+ kunder i over 70 lande Leaders Have Vision™ visionsolutions. com 3
Montreal, Toronto Rochester Follow The Sun Dorchester UK China SLC Japan France Irvine Waltham Indianapolis = development center Leaders Have Vision™ Plus: • Advanced Tools & Resources • Global Consulting Reach • Global Certified Partner Coverage = support center = both visionsolutions. com 4
Stort IBM POWER Partner netværk • Over 15 Power Business Partnere i Norden – – Flere af dem har over 10års partnerskab med Vision Udvider løbende vores partner netværk IBM Iptor Konehuone Ergo. Group Logica IBM Caral Data Iptor i. Comp Merit ATEA Essist Pulsen IBM Soft Design Leaders Have Vision™ visionsolutions. com 5
Er der nogen forskel på High Availibility og Diaster Recover på POWER? Hvad er der af HA/DR alternativer til Windows? Leaders Have Vision™ visionsolutions. com 6
Hvad skal vi med en HA løsning når. . Leaders Have Vision™ visionsolutions. com 7
Skovbrand i Australien Leaders Have Vision™ visionsolutions. com 8
Kæmpe oversvømmelse i Pakistan og US Leaders Have Vision™ visionsolutions. com 9
Jordskælv på Haiti og i Peru Leaders Have Vision™ visionsolutions. com 10
1% 9% 90% Natural Disaster System or Application Failure User Created Issues The definition of disaster is different for everyone! Leaders Have Vision™ visionsolutions. com 11
På et privat hospital i Oslo, blev filer slettet… …. ved et uheld over frokosten – 25. 000 patientjounaler blev slette Leaders Have Vision™ visionsolutions. com 12
I Gøteborg blev lastbilerne holdende… …. på grund af en korrupt fil – varer til en værdi af €€€€ kunne ikke leveres til tiden… Leaders Have Vision™ visionsolutions. com 13
I København, kontor var lukket i flere dage… …IT brugte, ikke kun weekenden på at migrere til nyt hardware Leaders Have Vision™ visionsolutions. com 14
I Helsinki, IT-chefen blev fyret… …hans HA/DR plan levede ikke op leverandørens krav Leaders Have Vision™ visionsolutions. com 15
Hvad hører vi i medierne Leaders Have Vision™ visionsolutions. com 16
Nedetid er nedetid - Findes der en undskyldning for at være nede? Brugerne er ofte ligeglade med begrundelse. - Hardware opgradering Release opgradering Daglig backup Data opdateringer m. m - Nedetid koster penge uanset om det er planlagt eller uplanlagt. Leaders Have Vision™ visionsolutions. com 17
Omkostningerne ved at være nede Revenue Productivity • Direct Loss Number of employees affected X hours out times X burdened hourly rate • Compensatory Payments • Lost Future Revenue • Billing Losses Financial Performance Damaged Reputation • Customers • Suppliers • Financial Markets • Banks • Business Partners Know your downtime costs per hour, day, week, etc. • Revenue Recognition • Cash Flow • Lost Discounts (A/P) • Payment Guarantees • Stock Price Other Expenses Temporary Employees, Equipment rental, overtime costs, extra shipping costs, travel expenses, legal obligations Leaders Have Vision™ visionsolutions. com 18
RPO / RTO • • RPO (Recovery Point Objective) – RPO refererer til mængden af data der potentielt kan gå tabt ved et nedbrud. ”ved tape backup – risiko for optil 24 timers datatab”. RTO (Recovery Time Objective) – RTO refererer til den tid data/applikation er tilgængelig igen. ”hvorlang tid tager det at genskabe data manuelt”? Vi kan være nede i 24 timer uden det betyder noget Leaders Have Vision™ Vi tester vores backup, så vi er helt rolige hvis vi får et server-nedbrud visionsolutions. com 19
Down-time Vi kan være nede i 24 timer uden det betyder noget 16 00 17 00 18 00 19 00 20 00 21 00 22 00 23 00 00 00 01 00 02 00 03 00 04 00 05 00 06 00 07 00 08 00 09 00 10 00 11 00 12 00 13 00 14 00 15 00 System går ned 17: 00 IBM kommer med nyt hardware Starter restore fra bånd Hardware går ned – reationstid IBM kommer med nyt hardware Klargøring og restore af backup I alt Manuel indtastning 1 time = 3 timer 2 timer 6 timer 8 timer 16 timer xx timer Konklusion: For at sikre oppetid på 24 timer skal datatab ikke være mere end 3 timer fra sidste backup Uden HA/DR er der ikke sikkerhed for hvornår data er tilgængelig igen Færdig restore fra bånd Starter manuel indtastning af mistet data Færdig med manuel indtastning Frustreret brugere og kunder Leaders Have Vision™ visionsolutions. com 20
Hvorfor har alle virksomheder så ikke en HA/DR løsning? Has the number of applications that demand HA increased or decreased at your firm in the past two years? Source: ITIC - 2010 High Availability & Virtualization Survey May 2010 Leaders Have Vision™ visionsolutions. com 21
Hvorfor så? What concerns your firm most about software and hardware products designed to increase uptime of your business applications? (Select ALL that apply) Source: 2010 ITIC High Availability and Virtualization Survey Leaders Have Vision™ visionsolutions. com 22
Forskellen mellem HA (High Availability) & DR (Disaster recovery) – Data skal være tilgængelig 24/7 med failover mulighed = HA – Fleksible – Let at administere – Let at teste – Data skal kunne anvendes på backupserveren • • – RPO (datatab) = 0 sek* RTO = 15 til 30 minutter (uanset om det er planlagt eller uplanlagt nedetid) Data kan være utilgængelig i x antal tid, med minimal datatab = DR – Flexsible – Let at administere • • RPO (datatab) = 10 - 15 minutter RTO = 1 - 24 timer eller mere *) ved synchronous replikering Leaders Have Vision™ visionsolutions. com 23
Lad os gå lidt tilbage i tiden Leaders Have Vision™ visionsolutions. com 24
Datasikkerhed i over 15 år 1996 1990 2006 2004 + Leaders Have Vision™ 2007 2005 2010 2008 + + = visionsolutions. com 25
Portfolio Of Leading Brands VISION® for Windows® VISION® for Linux® VISION® for AIX® VISION® for IBM i® High Availability -Any distance data replication -One button switchover & failover -Hardware independent -Maximum flexibility -Low management -Physical or virtual environments -Optimal RPO and RTO TM VISION® Services Disaster Recovery -Any distance data replication -User selectable recovery point CDP -Hardware independent -Maximum flexibility -Low management -Physical and virtual environments -Public or Private Cloud deployment Data and Systems Management -Change Data Capture CDC -Data transformation -Data integration -Data migration without down-time -Analyze, monitor, manage and optimize Leaders Have Vision™ Note: Branding Updates In Process visionsolutions. com 26
Tre stærke Brands til IBM i Take backup to the next level Addressing the needs of SMB Environments Simplified HA in a Complex World Leaders Have Vision™ visionsolutions. com 27
i. TERATM Availability 6 Ideal for SMB type environments The industry benchmark for self managing HA i. TERATM Virtual Role Swap Test (VRST) – While-active role swap tests i. TERATM Fully Automated Role Swap – one button role swap process i. TERATM Transport Journal – Fast & robust object replication i. TERATM Direct Apply – optimized performance for SMB i. TERATM Real-Time Self-Correction – spend less time managing i. TERATM Apply Accelerator – spend less time tuning i. TERATM ZZ Audit – No locking required on backup system i. TERATM Audit Command Console – centralized audit control i. TERATM RSR Monitor – continuously shows you are ready to role swap i. TERATM Cluster Host Monitor – single screen multi-node monitoring i. TERATM Alert – Flexible e-mail alerting system i. TERATM SAT Lite – Fast automated analysis of your environment i. TERATM MQ Support – Web. Sphere MQ messaging support i. TERATM CRGs – Manage separate application groups on same system i. TERATM Quick Sync – no need to use tape for initial synchronization Migrate While Active – migrate servers without down-time Local Languages – English, Italian… Sold globally - supported locally Over 2000 licenses sold to over 1000 customers Leaders Have Vision™ visionsolutions. com 28
Hvad siger IT-cheferne? What concerns your firm most about software and hardware products designed to increase uptime of your business applications? (Select ALL that apply) Source: 2010 ITIC High Availability and Virtualization Survey Leaders Have Vision™ visionsolutions. com 29
Hvorfor tester vi ikke vores HA? • • Tid (skal ofte ske i en weekend) Usikkerhed Flere personer skal ofte være involveret, fra forskellige afdelinger (Økonomi, lager, produktion, IT mfl. ) Kompliceret, besværligt at lave test, svært at overskue Leaders Have Vision™ visionsolutions. com 30
Data Migration Projects 1 Live users Test users 2 Before Migration Users are live on the old system and database on new system is empty Electronic Synchronization i. TERA HA has the ability to schedule and manage the electronic synchronization of large libraries overnight. ITERA HA then keeps the new system synchronized 3 Cut Over Test i. TERA HA then allows users to sign-on to the new system and verify that the applications are Cut Over Ready. If not then we synchronize any missing objects or libraries until users are happy. 4 Cut Over is done and users continue working on the new system. Live users Benefits ● Take the down time and risk out of data migration ● No need to courier tapes ● No need to work weekends ● Cut Over test guarantees success Leaders Have Vision™ Tip: Over 20% of i. TERA customers also use it for data migration projects visionsolutions. com 31
Portfolio Of Leading Brands VISION® for Windows® VISION® for Linux® VISION® for AIX® VISION® for IBM i® High Availability -Any distance data replication -One button switchover & failover -Hardware independent -Maximum flexibility -Low management -Physical or virtual environments -Optimal RPO and RTO TM VISION® Services Disaster Recovery -Any distance data replication -User selectable recovery point CDP -Hardware independent -Maximum flexibility -Low management -Physical and virtual environments -Public or Private Cloud deployment Data and Systems Management -Change Data Capture CDC -Data transformation -Data integration -Data migration without down-time -Analyze, monitor, manage and optimize Leaders Have Vision™ Note: Branding Updates In Process visionsolutions. com 32
MIMIX AVAILABILITY 7 Ideal for enterprise type environments The industry benchmark for enterprise HA MIMIX Dynamic Apply Highest apply workloads in the industry Professional, Enterprise, Global Licenses – Match your environment Unlimited Application Groups – Individually switchable applications Auto Object Registration – automatic configuration of new objects Advanced Audit & Repair – self correcting / autonomic features Dynamic Apply – parallel apply jobs maximize performance Zero Downtime Backup – users are not interrupted to perform tape backups Automated Switch – flexible switch, predictable RPO/RTO Vision Solutions Portal (VSP) – portal based graphical monitor 5250 Menus, Screens & Commands – advanced integration and customization MIMIX For MQ – Web. Sphere MQ messaging support MIMIX Director Optimization Utilities – monitor/manage/optimize IBM i system MIMIX Monitor Jobs – customizable event control MIMIX Promoter – minimize down time while re-organizing & copying MIMIX Library Name Mapping – allows library renaming & duplicating Optimization Extensive optimization Bi-Directional (active-active) Flow – advanced workload balancing utilities for IBM i environments built-in Vision Solutions Portal (VSP) Sophisticated multi-node graphical HA & DR monitor Full feature HA & DR Leaders Have Vision™ visionsolutions. com 33
Der er sket rigtig meget siden – OMS/ODS tiden 1. 2. Single Point of Switch 3. Historical Reporting of Procedures and Audits 4. Customizable Automation 5. Enhanced Apply Performance 6. Integrated Control of Hardware HA 7. Leaders Have Vision™ Enterprise Monitor and Control System Health Monitor (MIMIX Director) visionsolutions. com 34
Dynamic Apply Architecture Maximizes apply performance Cooperative Processing Allows efficient cooperative processing of apply workload in multprocessor and multi-core environments resulting in faster apply rates Self-Maintaining File Relationships Intelligently handles relationships between physical and logical files by assigning them to the same or appropriate apply sessions automatically Self-Maintaining Data Integrity Automatically maintains data integrity of replicated objects including operations such as creates, deletes, moves, and renames MIMIX 7 Dynamic Apply Takes advantage of multi-core & multi-processor environments Multiple apply sessions work in parallel Per journal Reader Leverage multi-processor machines In February 2010 IBM launched its latest Power 7 processor. The Power 7 chip has up to eight cores, with each core able to run four threads. A Power 7 chip can run 32 tasks (session) simultaneously, which is quadruple the number of cores on the older Power 6 chip. Leaders Have Vision™ visionsolutions. com 35
Remote Journal vs. Local Journal Advantages of RJ Auditing & Self-healing • Sends journal entries to the remote journal instead of externalized entries – – • – – Externalizing entries is an expensive operation High performance / Less CPU intensive – – This is more efficient in terms of processing overhead This is in addition to sending internal entries Local journaling cannot support synchronous journaling Supports remote journal filtering (V 7. 1 i/OS) – – – • Advanced Filtering Ability to filter before image information from RUB and RDL entries Ability to filter journal entries for non-replicated files and objects Note: IBM RJ filtering support introduced in V 7. 1 i/OS reduces that advantage Supports both synchronous and asynchronous nodes – • • Sends from deeper inside the communication stack (below MI) – • Advantages of LJ Can be used to significantly reduce the amount of data sent to the target system Eliminate need to send unnecessary before images, work file transactions, etc… Note: This is considered one of the biggest benefits of local journal based HA • Supports both Remote Journal & Local Journal – – MIMIX Professional for SMB customers MIMIX Enterprise for Enterprise class customers Supports secondary data that normally doesn’t fit in standard journal entry without elaborate packing / unpacking support – – – Leaders Have Vision™ For support of LOB data (SAP) For support of large IFS writes For support of incomplete replayable entries 7 visionsolutions. com 36
Portfolio Of Leading Brands VISION® for Windows® VISION® for Linux® VISION® for AIX® VISION® for IBM i® High Availability -Any distance data replication -One button switchover & failover -Hardware independent -Maximum flexibility -Low management -Physical or virtual environments -Optimal RPO and RTO TM VISION® Services Disaster Recovery -Any distance data replication -User selectable recovery point CDP -Hardware independent -Maximum flexibility -Low management -Physical and virtual environments -Public or Private Cloud deployment Data and Systems Management -Change Data Capture CDC -Data transformation -Data integration -Data migration without down-time -Analyze, monitor, manage and optimize Leaders Have Vision™ Note: Branding Updates In Process visionsolutions. com 37
MIMIX® Recover. Now 5. 1 Recover More Data Than With Tape Alone*(1) Take backups to the next level *(1) Typical RPO with tape alone backups is 24 hours. Typical RPO with Recover. Now is 15 -30 minutes Nightly backups to a local/remote disk*(2) Users remain on-line during nightly backups Recover. Now Storage Platform / Vault Data changes saved to disk between backups*(3) Allows recovery potentially right up to point of failure Or recover data to any historical point in time (CDP) Uses any TCP/IP disk storage platform (e. g. Windows) Data compressed on disk typically by 60% Automates journaling for data capture Easy to set up and manage Easy to restore data to a new system/LPAR Used also for data migration / test environments Use stand alone or to add CDP to Vision HA *(2) Optionally nightly backups can be to tape, compatible with BRMS *(3) Typically every 15 -30 mins. User configurabl interval Recover objects to any historical point in time Leaders Have Vision™ visionsolutions. com 38
Standard Tape Backup Typically Gives 24 Hour RPO Backup to Tape There is typically a 24 hour interval between tape backups This means there is typically a potential for 24 hours of data loss! Lost data must be re-keyed after the tape backup is restored 24 hours of lost data can take days or sometimes weeks to re-key! Nightly backups typically introduce down time Tip: Recovery Point Objective (RPO) with tape is the number of hours between tape backups Leaders Have Vision™ visionsolutions. com 39
MIMIX® Recover. Now Backup Typically Gives 15 -30 Minute RPO Backup to Disk Data changes are captured and ‘pushed-to-vault’ frequently between backups There is typically a 24 hour interval between disk*(1) backups Data changes are captured and pushed-to-disk between backups*(2) Users can remain on-line during the disk backups Recovery possible potential right up to point of failure Recovery also possible to any historical point in time (CDP) *(3) Tip: Recovery Point Objective (RPO) with Recover. Now is the number of minutes between ‘push-to-disk’ *(1) Backups can be to disk vault or optionally to tape, its compatible with BRMS *(2) Push-to-vault is always to vault at user configurable intervals *(3) CDP is continuous data protection Leaders Have Vision™ visionsolutions. com 40
MIMIX® Recover. Now Architecture Recover. Now Storage Platform / Vault Tip: Recover. Now is compatible with BRMS so nightly tape backups can still be done to tape using BRMS commands… Recover. Now adds ability to save changed data to disk vault in between backups Leaders Have Vision™ visionsolutions. com 41
Flexible Configuration Topologies Some Examples: Local Vault Production System i Hosted Vault Recover. Now Platform Production System i or LPAR Any TCP/IP Storage Platform Frequent transfer of data changes Recover. Now Platform Same buildin g Remote Vault Frequent transfer Production of data changes System i Any TCP/IP Storage Platform Recover. Now Platform Any TCP/IP Storage Platform External TCP/IP communications Leaders Have Vision™ Any distance Tip: A single Recover. Now vault can support multiple sources visionsolutions. com 42
Portfolio Of Leading Brands VISION® for Windows® VISION® for Linux® VISION® for AIX® VISION® for IBM i® High Availability -Any distance data replication -One button switchover & failover -Hardware independent -Maximum flexibility -Low management -Physical or virtual environments -Optimal RPO and RTO TM VISION® Services Disaster Recovery -Any distance data replication -User selectable recovery point CDP -Hardware independent -Maximum flexibility -Low management -Physical and virtual environments -Public or Private Cloud deployment Data and Systems Management -Change Data Capture CDC -Data transformation -Data integration -Data migration without down-time -Analyze, monitor, manage and optimize Leaders Have Vision™ Note: Branding Updates In Process visionsolutions. com 43
One Product Downtime • • Data Loss Geographic Distance Virtualization Real-time Replication – – Minimize data loss and downtime Replicate over existing WAN bandwidth Virtualized Failover – – Protect v. Sphere or Hyper-V environments with host-level management and real-time replication Failover P 2 V, V 2 P, P 2 P, or V 2 V Application Protection – Exchange, Share. Point, SQL Cluster Integration – – Create “Stretch Clusters” using the Geo. Cluster feature Replicate from clusters to standalone servers or clusters Leaders Have Vision™ Replication Engine Virtualized Failover Application Protection Cluster Integration visionsolutions. com 44
Technology Synergies – Deep Expertise P 2 P PHYSICAL P 2 V VIRTUAL P 2 C CLOUD Leaders Have Vision™ visionsolutions. com 45
Fleksibilitet Hardware spejling 1 -1 - Samme hardware på begge sider - Skal udvides samtidig - Meget lille fleksibilitet Site 1 Leaders Have Vision™ Site 2 visionsolutions. com 46
VISION® For Windows - Family Of Products full server or data only high availability – real-time any distance replication – bandwidth optimization – failover detection – application independent – MS clustering support –VMware support – the industry benchmark continuous byte-level disk-to-disk backup – SIS de-duplication – eliminates backup window – no manual backups – WAN optimization for remote backups – archiving – granular any point in time recovery continuous byte-level disk-to-disk backup to a remote hosted server Amazon web services provides the service – reasonable monthly fee – recover whole systems, whole database or single documents real-time DBMS level replication with transformation – capture data changes at column level – map and transform data to different target DBMS – Oracle – DB 2 – MS SQL Server – Sybase - Informix - Teradata hardware independent server and data migration – managed via a single wizard – minimize downtime – physical to physical – physical to virtual – virtual to virtual – storage to storage – site to site centralized management of diskless PCs –fast deployment – simplifies recovery – low cost storage – turns Windows server into industry standard i. SCSI SAN – individual desktop settings Leaders Have Vision™ visionsolutions. com 47
Hvad er det Vision og Visions partnere kan hjælpe med? • • Hjælpe med at finde de mest kritiske applikationer Få et billede af hvad nedetid koster Dine RPO/RTO krav Finde den rette løsning Leaders Have Vision™ visionsolutions. com 48
Besøg vores hjemmeside Leaders Have Vision™ visionsolutions. com 49
Tak Leaders Have Vision™ visionsolutions. com 50
Скачать презентацию GS Gruppen 7 juni 2011 High Availability Disaster
e4241cebd20f0fe78f94a3e233481a16.ppt