Storage Monkeys

This article was updated with the latest news on sept 2th at 15.50 (CET)

The story is ended: Dell didn’t respond to the latest call for the 3Par bid and HP officially won for a total price of 2.4B$! Indeed, the 3Par press release is very strange (read third paragraph) but analysts and bloggers all bet on HP for the win.

As you know, after a first official bid from Dell in the mid of august, we saw something like an ebay auction for 3Par where Dell and HP battled. Now the game has come close to an end and we may wonder about what this acquisition means:

The good for HP-Dell-3Par

• HP has a new best of breed technology in enterprise storage. They will own a strong, sexy and competitive product to sell in the high end enterprise storage segment and the technology to design a scaled down version to definitively kill the sluggish EVA storages.
• Dell didn’t spend 2B$ for 3Par: very costly for a new entry in the enterprise storage market segment and wouldn’t be easy to explain investors how to quickly monetize their investments! They did a great job with Equallogic but they have a big presence in SMB market with a consolidated sales model but we can’t say the same for the enterprise storage market.
• The good news for 3Par is the finale price: 2.4B$! I can say only WOW here.

The bad for HP-Dell-3Par

• Hitachi won’t wait to lose customers… they will propose directly to customers the evolution of their next generation USP (highly awaited) and some of them will remain with HDS.
• Dell shown clearly his plans to the world, what will do EMC in the near future? They will try to extend their presence in every Dell/EMC deal/account to diminish Dell, for sure.
• HP is famous for their acquisitions and a lot of questions comes to my mind: will 3Par engineers stay in HP? will the product be developed at its best? how much time HP will need to integrate 3Par in its offering? and other questions rising about EVA customers: one above all, will they wait for 3Par?

The ugly for HP-Dell-3Par

• HP owns an high-end product now and I don’t know if 3Par has a ready scaled down version of his products to stay in the low-end EVA-like market. If not, Engineering of new small models will take time and EVA customers will be heavily attacked by all other vendors in the meantime!
• Dell guys failed one of the most important acquisitions in their career, now they need to carefully plan their next moves in enterprise data storage (many people point out to Compellent) to enable their cloud computing strategy… or they risk to miss the train.
• 3Par employees will not have a great future in HP… especially sales and administrative people.

Un paio di giorni fa sono usciti i dati di vendita del secondo trimestre per i server elaborati da IDC (qui un articolo che ne parla).

Allego anche la tabella riportata nell’articolo menzionato con i risultati:

Cosa dire? i dati si commentano da soli:

• HP è sempre l’azienda che vende di più (x86 la fa da padrone ovviamente),
• IBM ha perso un po: soprattutto la parte Power mentre la richiesta per gli x86 continua ad essere abbastanza forte,
• Dell ha fatto una crescita impressionante: +36,5%!!!!
• Oracle, come tutti ci aspettavamo, ha fatto un’altro -6%. Qui i motivi sono chiari a tutti: SPARC allo sbando, politiche coercitive, impegno lato Solaris non chiaro, ecc., ecc. Spero che si schiariscano le idee prima che sia troppo tardi
• e poi c’è Fujitsu che ha fatto un dignitoso 7,9%: secondo me in mezzo ci sono anche un bel numero di vendite sul fronte SPARC…

Che dire? è chiaro che Dell sta facendo molto bene, me lo confermano anche i clienti soddisfatti, mentre continua il lento declino delle piattaforme proprietarie sotto la pressione delle CPU intel.

Sto seguendo una discussione su LinkedIn sulle impressioni d’uso che alcuni clienti hanno avuto sullo storage Compellent. Se sei curioso di conoscere cosa dicono i clienti rigurdo a Compellent ti consiglio di dare una occhiata qui.

Segnalo gli ultimi due commenti sulla discussione:

“We’ve been using Compellent since 2007. We’re a service provider that uses Compellent to provide on-demand storage within our data centers, and replication to others that have Compellent arrays in production for recovery purposes. Overall, we’ve been very pleased. Have three arrays in production today, across three data centers. We’ve got one more going into production for our IT organization (splitting them off of our current large shared array), and another going in production to provide dedicated infrastructure for a managed services client. Bumps happen. But how those bumps are handled iswhat really matters. Compellent has taken owner ship of any issue that has popped up, and provided support that has been second to none.”

“We’ve been Compellent customers for nearly 3 years now and have had a great experience. I purchased 4 Compellent SANs for different companies we manage plus a DR site. Having deployed and/or worked with Hitachi, HP and Netapp SANs I find that Compellent SANs are by far the easiest to setup, upgrade, and manage. I have too much to do to spend all my time managing a SAN, and I now spend about 3 hours a week managing 4 of them. Having a TAM and monthly health checks as a service provides value, is relatively cheap insurance, helps us get technical or architectural questions answered in a proactive manner (not that this was an issue prior to having a TAM), often by the engineers who code the software, and generally ensures that I maintain healthy systems. Compellent support is top notch, by far the best I’ve worked with so far. Also being the virtualization architect I hope to gain even greater efficiencies with Compellent over time, and look forward to seeing what new VMware-specific and other application-specific integrations or features will be announced at VMworld next week.”

La mia opinione è che sono commenti molto positivi, soprattutto riguardo il livello e la qualità dell’assistenza tecnica (Copilot Support).

Se sei un cliente Compellent ti suggerisco di iscriverti a questo gruppo su LinkedIn e condividere la tua esperienza con gli altri!

Aggiungo anche una breve lista di altri gruppi: Faccebook (official page e users group), LinkedIn (official e users group), Storagemonkeys group.

Unisciti a noi: è un modo molto interessante per condividere la conoscenza, le idee e le esperienze o, se non sei cliente Compellent, per avere la possibilità di valutare il prodotto dal punto di vista di chi lo sta usando sul campo!

I’m following a discussion on LinkedIn about very positives user experiences of Compellent stuff, if you are curious about what customers say about Compellent’s SAN get a look here.

the last two comments are saying:
“We’ve been using Compellent since 2007. We’re a service provider that uses Compellent to provide on-demand storage within our data centers, and replication to others that have Compellent arrays in production for recovery purposes. Overall, we’ve been very pleased. Have three arrays in production today, across three data centers. We’ve got one more going into production for our IT organization (splitting them off of our current large shared array), and another going in production to provide dedicated infrastructure for a managed services client. Bumps happen. But how those bumps are handled iswhat really matters. Compellent has taken owner ship of any issue that has popped up, and provided support that has been second to none.”

“We’ve been Compellent customers for nearly 3 years now and have had a great experience. I purchased 4 Compellent SANs for different companies we manage plus a DR site. Having deployed and/or worked with Hitachi, HP and Netapp SANs I find that Compellent SANs are by far the easiest to setup, upgrade, and manage. I have too much to do to spend all my time managing a SAN, and I now spend about 3 hours a week managing 4 of them. Having a TAM and monthly health checks as a service provides value, is relatively cheap insurance, helps us get technical or architectural questions answered in a proactive manner (not that this was an issue prior to having a TAM), often by the engineers who code the software, and generally ensures that I maintain healthy systems. Compellent support is top notch, by far the best I’ve worked with so far. Also being the virtualization architect I hope to gain even greater efficiencies with Compellent over time, and look forward to seeing what new VMware-specific and other application-specific integrations or features will be announced at VMworld next week.”

In my opinion these are awesome comments, especially about Copilot Support.

If you are a Compellent’s customer I would like to suggest you to join this LinkedIn group and share your experiences with the community!

Moreover, here is a list of other Compellent’s communities on Faccebook (official page and users group), LinkedIn (official and users groups), Storagemonkeys group.

Join us: it’s a great way to share knowledge, ideas and experiences or, if you are not a customer yet, to better understand the product from the user point of view!

Qualche giorno fa è successo quello che tutti sapevamo, Oracle ha ottenuto il risultato che attendeva da tempo: il board di OpenSolaris si è sciolto.

Sappiamo tutti di quanto amore è capace Oracle nei confronti dell’opensource e sappiamo anche quanto questo passaggio era stato predetto. Ma diciamoci la verità, forse non tutto il male viene per nuocere (spero): Alcune note che sono girate su internet danno l’idea che Oracle voglia solo avere più controllo sulla cosa (ma và?) cambiando il modello con cui viene rilasciato Opensolaris (che cambierebbe anche nome ripescando il vecchio Solaris Express).

Comunque andranno le cose, gli sviluppatori che portavano avanti il progetto OpenSolaris si sono organizzati per tempo e hanno sempre mantenuto e scaricato gli ultimi sorgenti fino a quando è stato possibile, è già disponibile in rete un fork di OpenSolaris che si chiama Illumos. Adesso sarà da vedere chi sposorizzerà l’iniziativa e quanto, anche se mi sembra di aver capito che Nexenta abbia già iniziato a contribuire e la comunità sta rispondendo egregiamente.

Yesterday we saw a big tweetstorm about 3Par acquisition from DELL or HP and, of course, the discussion mentioned the possible alternatives for the loser of this fight, but it is clear that Dell and HP are looking for an “enterprise storage” capable to compete with other Tier 1 arrays/vendors in the high end market segment and the only solution seems to be 3Par.

The community, as often happens, breaks up into many parties. In this post I would like to express my point of view and some basic concepts about the meaning of enterprise storage in 2010. Enterprise storage comes in many flavors according to the goals you want to achieve.

Once upon a time was the “Enterpise Storage”

In the last century the concept of enterprise storage was very tightened with the capability to connect to Mainframes and it was very simple to tell enterprise storage from non-enterpise storage. The ES was a monolithic object capable to deliver great performance on proprietary hardware and protocols and connected to a single host at a time.

Things slowly changed with the enterprise adoption of Unix first and Windows later. Moreover, SCSI (small computer system interface) was born and some fades between black and white appeared. In the ’90s the enterprise storage was still a Monolithic storage capable to connect many (for that age) different systems (MF comprised). We saw, for the first time, on unix and windows servers, some interesting features like clones and replicas.

In this age we found the first “non-enterprise” or modular arrays, the first SANs and a first approach to standard hardware (HDDs).

What is “enterprise/T1 array” today?

The capability to connect to MFs is no more important for many companies, they already moved to other platforms  years ago and the enterprise storage array is losing very fast the monolithic form too.

But there is something still of primary importance: the architecture.

The architecture makes the difference: many fast redundant controllers, many connectivity ports and protocols, smart and huge shared caches and switched backplanes are only some of the most important elements in this kind of architectures.

A T1 array is designed to support many different workloads simultaneously, to scale up vertically and linearly, to have no point of failures and a great uptime.

What is T1.5 storage?

The industry has defined the big enterprise arrays T1 but companies have more than primary data to access to, so T2 is what we commonly call mid-size or modular (enterprise) storage. Sometimes little bit of confusion arises because Tiers are often referred to when talking about products and vendors too: EMC/IBM/HDS/NetApp are commonly identified has T1 vendors while 3Par/Compellent/Xiotech/ecc. are often named as T2.

T2 arrays are architected in a simpler way than T1 ones: two controllers, smaller mirrored caches, less scalability and connectivity but (often) easy to use.

T2 arrays are chosen for their lower cost (if compared to the T1s), for specific applications/workloads, for testing or as primary storage in SMBs and many other times when you want to avoid the complexity and costs of T1 arrays.

But there is a new category: 1.5. what is 1.5? sometimes we have 1.5 vendors: we can find here some small but fast growing vendors with T1-arrays comparable features products. (probably it’s a diplomatic way to not diminish too much small vendors with good products).

We can also term 1.5 arrays, the most famous product in this category is XIV: IBM is selling it as a T1 array but the rest of the world consider it as a T2 array… and here you are the the solomonic 1.5!

The evolution of “enterprise” concept: modular, commodity, features, integration and automation.

What IT managers are looking for in 2010? simple: saving money while bringing more to the enterprise in terms of quality and services! (ok, I know, it’s easier to say than to do).

From the infrastructure point of view, virtualization and consolidation are a mainstream way to do this. Enterprises are looking deeper and deeper at features, capabilities and architectures to simplify and automate the management of data stored in their boxes and, on top of all, they need open and integrated solutions to get the best results.

Vendors like Compellent, 3Par, NetApp and others are in the spotlight because they address very well these needs (surely better than T1s) and this is why they are climbing sales charts furiously (also in Fortune 500).

The concept of enterprise storage is evolving quickly because nowdays:

• virtualization is more important than vertical scalability: you can buy more boxes if you have a simple way to manage them as a whole.
• automation is more important than big hardware: save and use better resources with automations without spending time to optimize manually.
• integration is more important than proprietary management tools: a true integration allows big time savings.
• …and so on.

Here is a few words recap: enterprises are looking more now than in the past at open, easy to use, features rich, virtualized and automated storage!

Bye, bye T1 arrays?

It’s not difficult to preview the next steps: Federation and commoditization.

• Federation: in a coarse way, the capability of T1.5/T2 arrays to be managed and seen as a whole one to achieve T1 performances, scalability and uptime.
• Commoditization: the use of standard hardware (and tons of software) to build all the components of the architecture and save a lot of money!

We will have the ability to cluster (federate) a bunch of T2 arrays and use them as a T1 or, in other words, T1 arrays will be clusters of T2 arrays.

What is enterprise storage for you?

Well, I’m back from holiday and my internet connection was almost non-existent there but last week I heard about the last good shot from DELL in the storage world, they are buying one of the best and innovative enterprise storage vendors: 3Par!

IMHO, I think that Brocade and CommVault will be two (very good) targets to complete the hardware stack and to start to offer their own data protection/archiving software. (I say this because, as a Dell enterprise partner, I’m a little bit biased, )

BTW, you can find first reactions from EMC (@chuckhollis) and 3Par (@3parfarley) bloggers here.

BREAKING NEWS: HP Proposes to Acquire 3PAR for $24 per Share in Cash !!!

Solaris è stato, per anni, un sistema operativo Unix di eccellenza e forse lo è ancora oggi. Nelle sue diverse versioni ha fatto la storia dell’informatica: sulle prime workstation RISC, sui primi server SMP, sui grandi server da 64 e più CPU, e negli ultimi anni, dopo una serie di incertezze, anche sui server x86.

Il lavoro che aveva fatto Sun per far certificare Solaris da tutti i big vendor è stato encomiabile, ma le cose sono cambiate e molto velocemente. Sun è stata acquisita da Oracle e Solaris, come molti altri gioielli di Sun, sta scivolando sempre più nell’abisso.

I vendor stanno abbandonando i contratti OEM e lo sviluppo del prodotto, sia nella sua incarnazione open che in quella commerciale, sta vedendo i momenti più bui di sempre… probabilmente gli ultimi. Per chi non ne fosse al corrente ecco qui alcuni link che dovrebbero dare l’idea di cosa sta succedendo: qui, qui, qui, qui (questi sono solo i primi che ho trovato: la lista è lunghissima).

A quanto sopra c’è da aggiungere la totale mancanza di ogni comunicazione sul futuro dei server basati su CPU SPARC: nessun annuncio in merito, CPU vecchie, nessuna roadmap, c’è solo l’oblio.

Perchè Oracle sta facendo questo? non lo so, devo ammetterlo, è fuori dalle mie capacità cognitive: aveva dei prodotti eccellenti, dei clienti che li adoravano e una comunità attenta allo sviluppo. Probabilmente dei ritorni da questo tipo di operazioni ci sono ma più che di un ritorno in termini di consolidamento industriale e maggior qualità per i clienti mi sembrano mere operazioni finanziarie.

Oracle sta comunque raccogliendo i contratti di assistenza e garantisce la manutenzione (a prezzi fuori misura e con degli SLA, a volte, ridicoli), non che questo sia di tanto conforto per i clienti ma almeno non dovranno migrare le loro applicazioni dall’oggi al domani… possono aspettare domani l’altro.

Il RAID 6 è un meccanismo di protezione dei dischi più sicuro del RAID 5 ma ancora più lento in termini di prestazioni. Il RAID 5 non è più all’altezza della situazione, questo è ormai anni che lo sentiamo dire, ma ormai con i dischi dell’ultima generazione (1 e 2TB) e i prossimi (3TB di prossimo annuncio) è veramente troppo pericoloso!

Un RAID group formato da dischi SATA da 1TB/7200RPM ha un tempo di ricostruzione, su un array tradizionale, che può facilmente superare le 24 ore! Questo tempo si allungherà ulteriormente e le previsioni non sono rosee: con i dischi da 3TB sarà di circa 96 ore. Sempre, fra l’alto, che l’array abbia degli algoritmi di ricostruzione efficienti e le IOPS richieste sul front-end non siano eccessive. Quanto tempo ti puoi permettere di rimanere senza una adeguata protezione?

La soluzione ci sarebbe, si chiama RAID 6, ma questa crea un problema di performance non indifferente. Già il RAID 5 subisce quella che in gergo si chiama “write penalty”. La penalità è dovuta al fatto che, per ogni operazione di scrittura verso l’array, il controller deve prendere il blocco, dividerlo per la lunghezza dello stripe, calcolarne la parità e, infine, scrivere tutte le parti sui dischi. Tutte queste operazioni vengono spesso fatte con un controller hardware che, per quanto veloce, ha dei limiti che spesso rallentano comunque le performance globali del sistema. Il RAID 6 non deve calcolare una parità ma due e quindi questa operazione è ancora più onerosa.

Il RAID 5 o il RAID 6 vengono spesso preferiti quando ci sono pochi dati in scrittura e molti in lettura (es. web server), in questi casi i tempi di risposta sono ottimi (se il RAID group non è degradato) ma, per i DB e le applicazioni molto “write intensive” vengono preferite ancora oggi configurazioni in RAID 1.

Le implementazioni di RAID 6 sono le più disparate e alcune eliminano, almeno in parte, la write penalty. Alcuni vendor, infatti, hanno messo a punto alcune sofisticazioni che posso attenuare il peso del calcolo della doppia parità.

NetApp, ad esempio, ha implementato una sua versione di RAID 6 che si chiama RAID-DP. Il RAID-DP è una evoluzione del RAID 4 (altro sistema di protezione simile al RAID 5 sempre utilizzato negli array NetApp). Il RAID-DP non aggiunge molto a quanto già conosciuto in precedenza con il RAID 4 ma, distribuendo i blocchi scritti in maniera differente da quanto fatto con il RAID a singola parità (diagonalmente) si ottiene un meccanismo molto più resiliente. Inoltre, l’implementazione specifica di questo particolare tipo di RAID 6 permette dei tempi di ricostruzioni molto rapidi.

Compellent, ha invece un approccio diverso e basato sulla sua funzionalità principe: la data progression. Il RAID 6 ha una implementazione più tradizionale (quindi lenta in scrittura) ma i singoli blocchi vengono sempre scritti in RAID 10 e poi convertiti, in modo asincrono e in background. Anche questo approccio risulta molto efficiente e garantisce al cliente adeguata protezione e performance.