Samsung and Seagate Unleash Killer SSDs

While some of the flash array vendors struggle to even get 3D NAND into their arrays, Samsung and Seagate have unleashed next generation SSDs that simply topples the scales of measurement of these devices.  Consider that some flash array vendors offer storage capacity of 35 TB in 3 rack units – Samsung is offering a 32 TB 2.5″ SSD and you get a feel for the magnitude of what is happening in this segment.  Of course there are other aspects to be considered but it is a stunning development that offers a number of happy alternatives to storage engineers.

Samsung put their 32 TB SSD into a 2.5″ form factor while Seagate’s is in a 3.5″ form factor.

You can read more on Samsung’s SSD here and here.

You can read more on Seagate’s SSD here.

Recommended Reading: Symantec/Intel Architecture Compares to Flash Array Architecture for Oracle Databases

It certainly is worth looking at the latest white paper from Symantec and Intel.  They have dropped a small bomb on the flash array party. In a white paper, Software-defined Storage at the Speed of Flash, the duo provide a look at a nice Oracle database architecture where they show both price/performance advantages and comparable performance to flash arrays from Violin Memory, EMC and a CISCO solution.  Two Intel R1 208WTTGS 1RU servers were outfitted with four Intel P3700 Series SSDs, 128 GB DDR4 Memory, Symantec Storage Foundation Cluster File System 6.2, Oracle 11gR2 and Red Hat Enterprise 6.5 OS. The two servers are interconnected with high speed dual-port Intel Ethernet Converged Network Adapter. The white paper goes into quite a bit of detail and offers a nice chart comparing the converged solution with the flash arrays solutions.screenshot_475

 

 

Recommended: Flash Accelerating Oracle RAC in 16RU & VMware Power Savings with Local Flash

Fusion-io, Dell, and Mellanox have built a very small footprint (16RU), extremely fast, and price-sensitive reference architecture for Oracle RAC 12c. It is very impressive on a number of counts.  It further highlights the move to using local flash storage over remote flash storage arrays.  This solution aims at reducing hardware sprawl and power consumption by achieving higher performance in a smaller footprint.  From the description – “This reference architecture features four Fusion ION Accelerators and four Oracle RAC database nodes all connected through redundant Mellanox SX6036 switches. Each ION Accelerator in this reference architecture uses a Dell PowerEdge R720 with three 56 Gbps Mellanox ConnectX-3 InfiniBand cards and four industry-leading Fusion ioDrive2 2.4TB flash storage devices for a total of 9.6TB of all flash storage. Each of these ION Accelerator units fits in 2 rack units (2RU) and delivers well over 645,000 8K database IOPS and up to 12 GB/s sustained throughput. Data redundancy is maintained across pairs of ION Accelerators using synchronous writes, thus providing system and data high availability.  The Oracle RAC nodes consist of Dell PowerEdge R620 servers. Each two-socket server consumes a mere 1U of rack space and yet is capable of pulling 1.4 million 8K IOPS from the ION Accelerator storage layer as measured by the Flexible IO Tester utility.  The Oracle RAC nodes are connected to the ION Accelerators through redundant Mellanox switches. Each Oracle RAC node sees the ION Accelerator storage as simple multipath block storage. The multipath devices are aggregated by Oracle ASM to create a large and powerful diskgroup.  Expanding the size and performance of the database is as easy as adding more ION Accelerator devices to the ASM disk group.”  The links to the article and the Reference Architecture follow.

FIO_OracleRAC_1

FIO_OracleRAC_2

Increasingly, software vendors  see the virtue of putting their flash storage locally on the server.  For example, VMware engineers wrote recently that by replacing array shared storage with PCIe flash card storage on the servers they could substantially reduce power consumption and maintain high levels of performance.

VMware_Power_01

 

 

OpenStack Announcement : SolidFire/Dell/Red Hat Unleash SolidFire Agile Infrastructure (Flash Storage-Based) Cloud Reference Architecture

Very, very cool announcement today from SolidFire, Dell and Redhat. Everyone interested in the next generation datacenter should listen to this excellent keynote from a SolidFire’s CEO.  They are way, way ahead of the rest of the flash storage crowd.  Unlike announcements of some vendors whose native operating systems don’t even support extreme scale-out and quality-of-service and that announce their belated participation in the OpenStack foundation minus virtually any substantial meat – SolidFire delivered a strong announcement – a real-world, pre-tested, pre-validated Dell/Redhat/SolidFire reference architecture for building a flash-based cloud. And they are not new to OpenStack they have been supporting it for some time. Today, SolidFire unleashed a pre-validated, pre-tested reference architecture with two key players – Dell and Redhat. In this talk – two eBay engineers – Subbu Allamaraju (Chief Engineer, Cloud) and John Brogan (Cloud Storage Engineering) discussed with Dave Wright (CEO, SolidFire)  the challenges that moved them to look at OpenStack and how eBay is using OpenStack today.  It is definitely worthwhile to listen to these knowledgeable  eBay engineers provide a meaningful discussion on why OpenStack is important.

Then, Dave Wright discusses the newly created OpenStack cloud reference architecture.  You can find the Solidfire Agile Infrastructure (AI) Reference Architecture  here :

soldfireAICloud

David goes into some of detail on the reference architecture for a scale-out cloud. Two Dell and Redhat execs also joined David on stage to further discuss the new reference architecture.

Recommended Reference Architecture : Scalable Flash Storage for Building a High Performance ESX Cluster

Kaminario has an excellent write-up on how they accelerated a mixed workload of SAP which was using Microsoft SQL Server on a VMware cluster. It is an excellent read if you are interested in boosting performance of your ESX cluster and the apps running within it.

kam01

You can read the full report :

kam02


gotostorageGo to more posts on storage and flash storage at http://digitalcld.com/cld/category/storage.


Recommended Reading: Exadata X3 – Measuring Smart Scan Efficiency With AWR

Ran across this on twitter. If you are interested in getting datapoints on Oracle Exadata there are some nice new data points from Trivadis that are certainly worth reading.  Keeping in mind that this is Exadata X3 (1/8 rack = 2 servers) and that Exadata X4 is now available. The report is in PDF format.

trivadis01

 

 

Recommended Reading : Systems Performance, SmartOS and the Cloud

There is a new book out, Systems Performance : Enterprise and the Cloud by Brenden Gregg, and if you have to put up with slow software, book_brendenthen on a Monday, you need this book.  If you don’t know who Brenden Gregg is, then you haven’t been paying attention to a revolution in some of he  software that came from his team at Sun just a few years back. And this book constitutes an excellent way at  looking at system performance.  As described on Amazon : “Large-scale enterprise, cloud, and virtualized computing systems have introduced serious performance challenges. Now, internationally renowned performance expert Brendan Gregg has brought together proven methodologies, tools, and metrics for analyzing and tuning even the most complex environments. Systems Performance: Enterprise and the Cloud focuses on Linux® and Unix® performance, while illuminating performance issues that are relevant to all operating systems. You’ll gain deep insight into how systems work and perform, and learn methodologies for analyzing and improving system and application performance. Gregg presents examples from bare-metal systems and virtualized cloud tenants running Linux-based Ubuntu®, Fedora®, CentOS, and the illumos-based Joyent® SmartOS™ and OmniTI OmniOS®. He systematically covers modern systems performance, including the “traditional” analysis of CPUs, memory, disks, and networks, and new areas including cloud computing and dynamic tracing. This book also helps you identify and fix the “unknown unknowns” of complex performance: bottlenecks that emerge from elements and interactions you were not aware of. The text concludes with a detailed case study, showing how a real cloud customer issue was analyzed from start to finish.”   You can get a quick overview of the book by looking at his presentation and for the first time being introduced to the DTrace pony (page 6, but we will come back to the DTrace pony later) you can experience modern performance tools.

gregg02

Any so begins your magical mystery tour if you don’t know DTrace. If you haven’t experienced it you have missed out on a tool capable of literally

DTracegiving you deep observability into your system.  Though the best versions of DTrace run on open-source Solaris-based operating systems like SmartOS and illumos-based variants, their are also versions that run on Linux.  Couple with DTrace are utilization and latency heat maps and flame graphs which offer a visual view of what DTrace is reporting. As you get more interested in this tool you may want to examine what people like Brendan Gregg and Joyent are doing with DTrace in the Cloud –

gregg05

Joyent is using a forked version of OpenSolaris based on the illumos (open source) version.  Joyent’s open source version, SmartOS,  which they have customized around cloud deployments provides a large number of features unavailable on Oracle’s version of Solaris. And because SmartOS leverages OpenSolaris, one more thing becomes useful, ZFS, one of the most sophisticated file systems available.

gregg06

Another virtualization aspect of SmartOS is that it leverages Solaris zones (extremely lightweight) and Linux KVM (ported, debugged and tuned via DTrace).  And this all has given Joyent very special insight into dense virtualization and the architectures that virtualized clouds are built on.

gregg08

All of this is very interesting and offers an alternative to the usual VMware cloud, it also offers very strong open source enterprise-tested features of virtualization, file systems, tracing software and a suite of great features both server, storage and software-centric.  You can learn more about Joyent’s cloud version of Solaris at SmartOS.org.

Recommended Reading : MariaDB 10.0 Beta

It’s worth noting that when Oracle bought Sun Microsystems, one of the big concerns was Oracle’s stewardship of the open source MySQL database. Michael Widenius spent tremendous energies trying to exclude the MySQL from the deal.  In the end, Oracle ended up with MySQL, but that didn’t prevent Widenius from forking MySQL, creating MariaDB and today taking away large swaths of business and mind-share from MySQL.  As many businesses including Google move to MariaDB – the question is whether it is over for Oracle MySQL ?  It is an open question.  There is certainly a large amount of suspicion with regards to Oracle’s motives with MySQL.  Now, Widenius’ team are releasing the latest and most feature-rich version of MariaDB to date.  Considering the Sun-Oracle deal closed in 2010 – they have moved fast and are offering what could be a serious competitive challenge to MySQL.  Consider that Red Hat is also switching its storage engine in Fedora 19 making MariaDB the default implementation of MySQL. Others like Mozilla are migrating to it.  Adoption by these large and influential companies set a trajectory for MariaDB that can only be described as ascendent.  The latest 10.0 release of MariaDB offers up feature after feature that is  absent in Oracle’s version of MySQL.  Widenius’ company,  skySQL just received $20 million Series B funding to add more fuel to the MariaDB growth. The focus of accelerating and providing further differentiation with Oracle’s MySQL is clearly a focus as MariaDB introduced support for Fusion-IO’s “atomic writes” feature earlier this year and as more features including global GIS support and improved master-slave replication.  With 10.0 there are a slew of features that further differentiates MariaDB from Oracle MySQL. In 10.0 there is over 1.5 million lines of new code.   Key new features around Parallel slave technology, parallel replication, integration with storage engines like Cassandra, Connect, Sequence, Spider and TokuDB.  It also has new ease-of-management features. Details of the release can be found in the MariaDB Press Release. You can see some of these in this presentation (slides) :

and in this benchmark you get a feel for the performance versus Oracle MySQL :

mariadb

Widenius has mentioned that the skySQL now has over 300 customers including  large enterprise customers including Deutsche Telecom, Craigslist, the Financial Times, Home Depot and many others.

The question is can Oracle MySQL compete with MariaDB 10.0 ?  It will certainly require time for customers to migrate, so for the moment Oracle may have some breathing space – but without a direct software response to MariaDB that is not enough. Customers that have or are migrating are laying down both a technical path of best practices to migration and perception of the situation that is not favorable to Oracle MySQL in the long run.   Widenius feels there is no longer a need to use Oracle MySQL and the skySQL website demonstrates this with a nice competitive matrix comparing MySQL 5.6 and MariaDB 10.0 :

mariadb02