Datacenter Consolidation
Transcription
Datacenter Consolidation
Sébastien CHENE Datacenter Solution Architect Geneva Business Center 12, avenue des Morgines 1213 Petit-Lancy 1 www.stinco.com [email protected] +41 (0)79 705 97 32 EMC Forum 2011 EMC Forum 2011 Devenez le héros de la sécurisation des données de votre entreprise • La Technologie: Déduplication • EMC Data Domain • Mise en œuvre de la technologie – Sauvegardes – Archivage – Stockage Multi-Tiers • Votre Partenaire: STINCO EMC Forum 2011 La Technologie: Déduplication Dramatically Reduces Storage Capacity Requirements 10–30 times less data stored versus fulls + incrementals with typical retention policies Data Stored 30 20 10 0 1 5 10 15 20 Weeks in Use Deduplication storage Traditional storage EMC Forum 2011 La Technologie: Déduplication Store more backups in a smaller footprint Friday Full Backup A B C D A E F Mon Incremental A B H Tues Incremental C B I E Weds Incremental Thurs Incremental A G G Backup Data Estimated Reduction Logical Physical FRIDAY FULL 1 TB 2–4x 250 GB Monday Incremental 100 GB 7–10x 10 GB Tuesday Incremental 100 GB 7–10x 10 GB Wednesday Incremental 100 GB 7–10x 10 GB Thursday Incremental 100 GB 7–10x 10 GB Second FRIDAY FULL 1 TB 50–60x 18 GB 2.4 TB 7.8x 308 GB J C K Second Friday Full Backup B C D E F L G A B C D E F G H I J K L EMC Forum 2011 H TOTAL Methodology: Inline vs. Post-Process Deduplication INLINE POST-PROCESS Deduplication Before Storing Deduplication Deduplication After Storing Store Deduplication 3x disk accesses to shared store Other activities unimpeded − Predictable − Simpler The more processes, the more resource contention − − − − Copy to tape: Too slow to stream tape Recovery: Service level agreement predictability Replication: Poor time-to-disaster-recovery Deduplication: If interleaved with backup or restore More administration to fight these issues EMC Forum 2011 Performance: CPU-Centric vs. Spindle-Bound Data Domain Throughput MB/s 6,000 Fibre Channel SATA Most deduplication vendors 50 50 100 Number of Disk Spindles EMC Forum 2011 150 200 Backup Data Reduction/Deduplication Time Series of Large Enterprise Implementation 2H '07 2H '08 1H '09 2H '09 1H '10 1H '11 In Use Now 15% 15% 24% 14% 12% 27% 8% 40% 31% 16% 46% 14% 22% 14% 7% In Near-term Plan 21% In last three years, in-use rates for 25% backup with deduplication 26% have risen from 15% to 48% 6% 48% In Pilot/Evaluation 28% 15% 4% 25% 16% In Long-term Plan 20% 17% 7% 18% 10% 13% Past Long-term Plan Source: Wave 15 Storage Study – Q2 2011, published 5/16/11, large-enterprise sample; H ‘07, n=151; 2H ‘08, n=127; 1H ‘09, n=147; 2H ‘09, n=182; 1H ‘10, n=146; 1H ‘11, n=31;TheInfoPro (www.theinfopro.com) EMC Forum 2011 Not in Plan Purpose-Built Backup Appliances Open Systems + Mainframe EMC IBM HP Oracle EMC: 64.2% Quantum Sepaton FalconStor Dell Others Source: Worldwide Purpose-Built Backup Appliance 2011–2015 Forecast and 2010 Vendor Shares, May 2011, IDC. Chart: Worldwide Supplier Revenue, Total PBBA Market EMC Forum 2011 Devenez le héros de la sécurisation des données de votre entreprise • La Technologie: Déduplication • EMC Data Domain • Mise en œuvre de la technologie – Sauvegardes – Archivage – Stockage Multi-Tiers • Votre Partenaire: STINCO EMC Forum 2011 EMC Data Domain: Leadership and Innovation An history of industry firsts … 2003 2004 2005 First deduplication NAS First deduplication volume replication 2006 2007 First deduplication virtual tape library 2008 Largest deduplication array Fastest backup controller First deduplication directory replication First deduplication nearline storage STINCO Portfolio EMC Forum 2011 2009 Cascaded replication EMC Aquisition 2010 2011 First long-term retention system for backup and archive First distributed processing Data Integrity: Data Invulnerability Architecture End-to-end data verification Checksum Deduplication, write to disk Verify Self-healing file system Cleaning Expired data Defrag Verify Generate Checksum Verify Data File System Deduplication Local Compression RAID 6 Other RAID 6 NVRAM Snapshots EMC Forum 2011 End-to-end data verification Verify the file system metadata integrity Verify user data integrity Verify stripe integrity DD Boost Software • Distributes parts of deduplication process to backup server or application clients DD Boost • Licensable software works across Data Domain portfolio • Supports majority of backup software market • EMC Avamar and NetWorker • Symantec NetBackup and Backup Exec • Speeds backups by up to 50 percent • Process more backups with existing resources • 20–40% less overall impact to backup server • 80–99% less LAN bandwidth • Enables Data Domain replication management from the backup application EMC Forum 2011 Additional Data Domain Software Options Data Domain Virtual Tape Library Data Domain Replicator • Easily integrates with Fibre Channel • Network-efficient and encrypted • Emulates multiple tape libraries • Transfers only compressed, deduplicated data over the WAN • Supports open systems and IBM i operating environments • Consolidate up to 270 remote sites into a single system Data Domain Retention Lock Data Domain Encryption • File locking to satisfy IT governance and compliance policies • Inline encryption of data at rest • Electronic data shredding • Satisfies internal governance rules and compliance regulations • Protects against theft or loss of a physical system EMC Forum 2011 Network-Efficient Replication for True Disaster Recovery Lowers WAN costs; improves service level agreements 1–5% DB Flexible replication One-to-many Many-to-one Bi-directional System-to-system Cascaded Data Domain system Archive data Backup data Data Domain system 1–5% 1–5% Home Data Domain system Home WAN Data Domain Global Deduplication Array Source: Remote sites 95–99% cross-site bandwidth reduction EMC Forum 2011 Destination: Data Center Hub Supports hundreds of remote sites DD Archiver Overview Cost-optimized, long-term retention • Data Domain system for backup and archive • Active tier: short-term data protection; less than 90 days • Archive tier: scalable long-term retention; multiple years • High-throughput deduplication storage • Up to 9.8 TB/hr • Cost optimized for long-term retention • Up to 570 TB usable, 28.5 PB logical capacity • Low cost per gigabyte while maintaining high throughput • Fault isolation of archive units for long-term recoverability • Leverage existing Data Domain system advantages • Supports DD Replicator and DD Retention Lock software • Data Invulnerability Architecture to ensure data integrity EMC Forum 2011 Data Domain Systems Trajectory Data Domain SISL Scaling Architecture: CPU-centric Improvement since 2004: Throughput: ~175x Capacity: ~450x Throughput GB/s 5 2014 (est.) 3 1.5 0.04 EMC Forum 2011 DD200 (2004) 2004 2010 2011 Future Industry’s Most Scalable Inline Deduplication Systems Global Deduplication Array DD800 Appliance Series DD Archiver DD600 Appliance Series Software options: DD Boost, DD Virtual Tape Library, DD Replicator, DD Retention Lock, and DD Encryption DD160 Appliance DD160 DD620 DD640 DD670 DD860 DD890 Global Deduplication Array DD Archiver Speed (DD Boost) 1.1 TB/hr 2.4 TB/hr 3.4 TB/hr 5.4 TB/hr 9.8 TB/hr 14.7 TB/hr 26.3 TB/hr 9.8 TB/hr Speed (other) 667 GB/hr 1.1 TB/hr 2.3 TB/hr 3.6 TB/hr 5.1 TB/hr 8.1 TB/hr 10.7 TB/hr 4.3 TB/hr Logical capacity 40–195 TB 83–415 TB 0.32–1.6 PB 0.6–2.7 PB 1.4–7.1 PB 2.9–14.2 PB 5.7–28.5 PB 5.7–28.5 PB Usable capacity Up to 3.98 TB Up to 8.3 TB Up to 32.2 TB Up to 55.9 TB Up to 142 TB Up to 285 TB Up to 570 TB Up to 570 TB EMC Forum 2011 With Data Domain Deduplication Storage Systems, You Can… Retain longer Keep backups onsite longer with less disk for fast, reliable restores, and eliminate the use of tape for operational recovery Replicate smarter WAN Move only deduplicated data over existing networks with up to 99% bandwidth efficiency for cost-effective disaster recovery Recover reliably Continuous fault detection and self-healing ensure data recoverability to meet service level agreements EMC Forum 2011 Devenez le héros de la sécurisation des données de votre entreprise • La Technologie: Déduplication • EMC Data Domain • Mise en œuvre de la technologie – Sauvegardes – Archivage – Stockage Multi-Tiers • Votre Partenaire: STINCO EMC Forum 2011 Cas Pratiques… Sauvegardes Environnement TPE – SMB - Branch Shared Storage NFS / iSCSI DD160 – DD620 WAN Optimized Replication NFS Datastore for backup CLOUD Service Provider OR DataCenter Consolidation vSphere Essential Up to 30 VMs EMC Forum 2011 Cas Pratiques… Sauvegardes Environnement SMB - Enterprise DD620 – DD670 Shared Storage NFS / iSCSI / FC WAN Optimized Replication VTL for AS400 NFS / CIFS, Ge or 10Ge CLOUD Service Provider OR DataCenter Consolidation AS400 BRMS vSphere Standard vSphere Enterprise EMC Forum 2011 Backup Server Physical Environment Unix / Linuxx / MS … Cas Pratiques… Sauvegardes Environnement NAS DD620 – DD670 WAN Optimized Replication NDMP VTL CLOUD Service Provider OR DataCenter Consolidation NAS NDMP NFS / CIFS Backup Server EMC Forum 2011 … Cas Pratiques… Sauvegardes Database Protection & Files Transfers DD620 – DD670 Shared Storage NFS / iSCSI / FC WAN Optimized Replication CIFS / NFS, Ge or 10Ge DB DUMP SQL, Oracle DB2, Sybase, … VM Transfer WAN optimized DB Server EMC Forum 2011 Pear DataCenter OR DataCenter Consolidation Devenez le héros de la sécurisation des données de votre entreprise • La Technologie: Déduplication • EMC Data Domain • Mise en œuvre de la technologie – Sauvegardes – Archivage – Stockage Multi-Tiers • Votre Partenaire: STINCO EMC Forum 2011 ARCHIVE Differents archives needs Economic Need: Manage exponential data growth Activities: • Automate transfer of fixed-content data from primary disk to archive • Index archived information • Create a stub on primary storage • Delete archived data from primary after transfer •De-duplicate redundant data stored in the archive Benefits: • Reduce backup window • Save on primary storage (Space & Cost) • Save on archive storage • Increase performance of primary and backup systems • Easily retrieve information EMC Forum 2011 Patrimonial Need: Long term retention of business critical or historical data Activities: • Transfer major company assets in long term storage area • Index archived information • Setup policies to manage data lifecycle based on its value Benefits: • Fast & easy access to the data • Preserve important information for long periods of time • Manage data lifecycle Compliance Need: Comply with regulations and enable ediscovery in support of litigation requirements Activities: • Setup automatic policies to archive data based on legal requirements • Automatically delete data • Control user access • Index archived information Benefits: • Comply with specific regulations • Easily retrieve data • Enable quick access to information in case of litigation What is used to do? Backup AND Archive Primary Data Backup Data Passive Archive • Archives is kept as a backup extension • Old backups are renamed “archives” • The challenges: – – – – – EMC Forum 2011 Primary Storage Explosion (Costly) Backup windows explosion (Degrade operations) Hard to access archived (Off-Line data) Searching unformatted information is a mess (no index) Store archived data on tapes: Reusability? Backup OR Archive? Different usages Different storage needs Different processes Ordering data in a structured and indexed way for long term conservation Automatically copying information in case files get lost or deteriorated ARCHIVING BACKUP Target: Data Preservation & Retrieval Target: Data Recovery Primary copy of fixed content data Create copy of dynamic data Fixed content kept for future reference Content is periodically overwritten Activity for long-term retention Activity for short-term retention Used for date retrieval and Compliance Used for Recovery purposes EMC Forum 2011 What’s going-on? Archive THEN Backup Backup Data Primary Data Active Archive • Fixed content: Final version of the data • Unchanged data moved from primary storage to archive: – Reduce Capacity on primary storage • Less “real” production data • Shorten backup window, improve backup and restore • Improve primary servers performance and users access – Index all archived contents • Easiest researches • Fastest retrievals EMC Forum 2011 Data Explosion Containment Managed Datas Use Case • 3TB capacity in 2010 • 30% yearly growth (Gartner 2009) • 2/3 to be archived (Atempo 2009) 20 TB Secured & Archived Data 10 TB 13,8 TB 27,6 TB 30 TB Weekly or Monthly archived Backuped Datas Containement Daily Backup 41,36 TB 40 TB 3TB Years EMC Forum 2011 Cas Pratiques… DD Archiver Cost-optimized, long-term retention • Data Domain system for backup and archive • Active tier: short-term data protection; less than 90 days • Archive tier: scalable long-term retention; multiple years • High-throughput deduplication storage • Up to 9.8 TB/hr • Cost optimized for long-term retention • Up to 570 TB usable, 28.5 PB logical capacity • Low cost per gigabyte while maintaining high throughput • Fault isolation of archive units for long-term recoverability • Leverage existing Data Domain system advantages • Supports DD Replicator and DD Retention Lock software • Data Invulnerability Architecture to ensure data integrity EMC Forum 2011 Cas Pratiques… Archives Free-up space: Files & Mails Backup Server BACKUP WAN Optimized Replication Archive BACKUP CIFS / NFS, gigabit ethernet Data Transfert EMC Forum 2011 CLOUD Service Provider OR DataCenter Consolidation Data Transfert Stub Mail Server DR Archive Stub Archive Server Mails Archive Server Mails NAS / File Server Cas Pratiques… Archives Email Legacy Archiving Retention Lock Retention Lock WAN Optimized Replication Archive CIFS / NFS, gigabit ethernet CLOUD Service Provider OR DataCenter Consolidation Data Transfert Stub Mail Server EMC Forum 2011 DR Archive Archive Server Mails Enterprise Vault Devenez le héros de la sécurisation des données de votre entreprise • La Technologie: Déduplication • EMC Data Domain • Mise en œuvre de la technologie – Sauvegardes – Archivage – Stockage Multi-Tiers • Votre Partenaire: STINCO EMC Forum 2011 Cas Pratiques… Stockage Multi-Tiers Comment gérer vos données non structurée? Applications and Users Inflexible Complex Inefficient Expensive EMC Forum 2011 NAS and File Servers Cas Pratiques… Stockage Multi-Tiers Enabling a Dynamic Storage Infrastructure Applications and Users Dynamic Seamless ARX File Virtualization Efficient Integrated EMC Forum 2011 NAS and File Servers Cas Pratiques… Stockage Multi-Tiers Comment gérer vos données non structurée? Applications and Users Global Namespace • Federates and presents logical representation of underlying file systems • Decouples access from physical location • Masks changes to underlying storage systems from applications and users Automated Data Management Policies • Automate common storage management tasks • Data migration • Storage tiering • Capacity balancing • NAS and File Servers EMC Forum 2011 Tasks performed without affecting access to files or requiring client re-configuration Cas Pratiques… Stockage Multi-Tiers Users Users mount a virtual CIFS share or NFS export presented by the ARX device 1 2 To users, files appear to reside in the virtual share CIFS / NFS Files actually reside in a physical share presented by a file server or NAS device CIFS / NFS 4 3 5 TIER 1 TIER 2 MOVE SSD / SAS $$$ TIER 3 MOVE SATA $$ BACKUP BACKUP EMC Forum 2011 $ The ARX proxies CIFS and NFS file access to the appropriate physical location The ARX can also automatically move or place files based on customized policy DR Site Cas Pratiques… Stockage Multi-Tiers F5 ARX: Real automated ILM enabler. • Global Namespace – Decouple logical file access from physical file location – Federate multiple storage devices and file systems • Non-disruptive Data Migrations Through Automated Policies • Manage movement of files • Place and move files based on value • Automated policies manage optimal placement of files • Advantages – Customize backup policies by tier – Break up large file systems EMC Forum 2011 Devenez le héros de la sécurisation des données de votre entreprise • La Technologie: Déduplication • EMC Data Domain • Mise en œuvre de la technologie – Sauvegardes – Archivage – Stockage Multi-Tiers • Votre Partenaire: STINCO EMC Forum 2011 Votre Partenaire Portfolio Services Conseils et Support en Projets Datacenter • Elaboration de cahier des charges – Rédaction d’appel d’offres - Dépouillement des réponses - Proposition d’acteurs clés - Aide à la décision Architecture et Gestion Infrastructures Datacenter • Architecture de centre de calcul – Audits, Etudes, Analyses, Conseils - Offres complexes: matériels, logiciels & services - Adéquation IT aux contraintes business • Gestion de projets complexes – Interlocuteur unique - Garantie du fonctionnement opérationnel de la solution - Gestion, planification des différents intervenants - Implication dans le projet jusqu’à la recette finale EMC Forum 2011 Votre Partenaire Portfolio Solutions Business Continuity Datacenter Consolidation Data Management Desktop Consolidation Backup and Recovery Servers & Storage Virtualisation Tiered Storage Architecture Workstation Protection Disaster Recovery Planning Branch Office Consolidation Information Lifecycle Management Workstation Virtualisation High Availability Green Technologies Legacy Archiving Client Approach EMC Forum 2011 Thin, Thick, Laptops Votre Partenaire Constructeurs & Editeurs Stratégiques EMC Forum 2011 Votre Partenaire Quelques références EMC Forum 2011 Data Domain Infrastructure and Ecosystem Supports a variety of workloads and data types VMware Microsoft Microsoft SharePoint Oracle SAP Backup Archive NAS, SAN, DAS CA HP Vizioncore IBM i EMC Bus-Tech EMC F5 Networks Symantec Atempo IBM Atempo BakBone Network EMC Forum 2011 Primary storage Archive Applications Backup Applications EMC Symantec CommVault Midrange and Mainframe Disaster Recovery Replication over WAN Merci. Sébastien CHENE Datacenter Solution Architect Geneva Business Center 12, avenue des Morgines 1213 Petit-Lancy 1 www.stinco.com [email protected] +41 (0)79 705 97 32 EMC Forum 2011