Monday, August 24, 2015

Coming Soon: SPARC M7 and Solaris 12

[SPARC M7 Die, Courtesy The Register]

Coming Soon: SPARC M7 and Solaris 12

Abstract:

Operating Systems and Software Vendors continue to struggle with the difference between 32 bit and 64 bit architectures, but the SPARC family of processors continues to roll out 64 bit CPU chips for data flagship Solaris 64 bit Operating System. Watching companies announce new products ahead of time is tricky because of Government Regulation, but sometimes watching less overt routes can provide a great level of insight as to what is coming soon.

[SPARC and Solaris Public Roadmap, courtesy Oracle Corporation]

Roadmap: Foretelling the Future

Oracle has a history of releasing public road maps for SPARC and Solaris. They have been fairly accurate, since Oracle acquired Sun Microsystems. The roadmaps are subject to change, but they give the Architect a good idea of what is coming and how to plan for it. As of August of 2015, Oracle's public roadmap indicates that a new SPARC is in Test and Solaris 12 is coming early next year. The naming for that SPARC processor is not clear in the image, but

[System Controller and Console image, courtesy Oracle]

Firmware: What's in the Wild

SPARC M7 is operational!

A recent firmware release indicates the following bug numbers have been resolved:

20246063 Kernel Zone panics on first boot on M7-8
20003379 New FRUID enums for 4S variants of M7-8 chassis
19336643 Internal error: Oops: 17 [#1] on M7-8 with SysFW build_33

The new SPARC M7, more than a dream, appears to be a reality. Understanding Oracle's naming conventions, the firmware notes indicates support for an 8 socket chassis. The Chassis and Processor clearly exists.

[Sun Microsystems Solaris Logo]

Solaris 12 is running under OpenBoot!

A recent firmware release indicates the following bug numbers have been resolved:

9485526 obp assumes an executable heap and fails on Solaris 12

Solaris 12 is apparently running on existing SPARC platforms at Oracle, which is a good sign!


[San Francisco Bridge & City, courtesy Oracle Corporation]

JavaOne and OpenWorld 2015

The place to be will be Oracle OpenWorld in San Francisco, California during October 25-29.OpenWorld will offer many seminars to attend, but there is also JavaOne! Register & Fly to one of the most anticipated conferences of 2015. Why is this so anticipated?

Seminars & Sessions at OpenWorld & JavaOne 2015

How big can you get? If you are struggling for performance: SPARC, Solaris, and Java can fix it.
Operating a 16-Terabyte JVM...and Living to Tell the Tale [CON1855]
Antoine CHAMBILLE, Global Head of Research and Development, Quartet Financial Systems
Amir Javanshir, Principal software engineer, Oracle
Is there a limit to the size of the heap the Java Virtual Machine can handle? Java blogs often report 100 GB as the maximum amount of memory the JVM can manage effortlessly. Yet Quartet FS develops ActivePivot, an in-memory analytic database written in pure Java that is frequently deployed in a terabyte of memory. In fact, earlier this year, it partnered with Oracle to deploy a large credit risk use case in 16 TB of memory and the 384 cores of an M6-32 SPARC server from Oracle. Yes, you can do it, once all the layers of the solution come together: data structures and thread management in the Java code, garbage collection in the JVM, memory management in the OS. This presentation shares all the steps to achieve this.
Conference Session
Do you use Oracle RDBMS? It runs fastest in Oracle silicon, but WHAT silicon? M7 is a good bet!
Top 10 Tips and Tricks to Run Oracle Database Best on Oracle Solaris Systems [CON2742]
Ken Kutzer, Principal Product Manager, Oracle
Viraj Nr, Principal Software Engineer, Oracle
Oracle Engineering has focused on the codevelopment of Oracle Database, Oracle Solaris, microprocessors, and Oracle Storage Cloud, resulting in significant advantages for customers who deploy these products together. With each new release, deeper integration results in breakthrough enhancements in security, performance, and ease of management. This session outlines these enhanced capabilities and how they benefit you. Learn the top tips on how to optimize your configurations for the best-possible results. Areas covered include tuning tips for reliable performance and faster database startup with Oracle’s revolutionary new Software in Silicon database features.
Conference Session
Security: Solaris under SPARC M7 Processor making Buffer Overflows obsolete.
Learning to Use SPARC M7 Application Data Integrity to Detect Buffer Overflow Attacks [HOL5447]
Claude Teissedre, Market Development Engineer, Oracle
Programming errors or input data checking inadequacies are vulnerabilities that can be exploited by attackers to alter the behavior of programs, and the “linear buffer overflow” attack remains a major threat to many applications today. Application Data Integrity (ADI), a Software in Silicon feature powered by Oracle’s SPARC M7 chips, allows the CPU to detect such memory corruptions at almost zero cost. This hands-on lab is primarily aimed at detecting simple attack patterns using the Oracle Solaris memory allocators and the Oracle Solaris Studio Discover ADI features. In this lab, advanced developers learn how to use the Application Data Integrity API to adapt a memory allocator to the ADI technology and to create a user and signal handler to customize the ADI error handling behavior.
HOL (Hands-on Lab) Session
 Database: More of Oracle RDBMS is executed in Silicon under Solaris with SPARC M7 Processor.
Speeding Up Oracle Database Using SPARC M7 Hardware Acceleration [HOL6011]
Adina Kalin-florescu, ISVE engineer, Oracle
Ling-yun Li, Principal Software Engineer, Oracle
Wen-sheng Liu, Oracle
Angelo Rajadurai, Technology Lead Oracle System, Oracle
Data Analytics Accelerator (DAX) is a Software in Silicon feature built into Oracle’s SPARC M7 chips. Eight DAX database offload engines are present on the SPARC M7 chip in addition to 32 cores. Many of the Oracle Database In-Memory 12c functions can run natively on the DAX, freeing the cores to do other work. Oracle has been integrating hardware and software together to give our customers the best platform for running Oracle Database. This lab looks at the advantage of Oracle Database In-Memory 12c on the SPARC M7 chip.
HOL (Hands-on Lab) Session
Next Generation: What is coming on Oracle's Roadmap. Solaris 12 will probably be there.
General Session: Oracle Solaris Strategy, Engineering Insights, and Roadmap [GEN8606]
Markus Flierl, Vice President, Oracle
This session discusses the strategy and roadmap for Oracle Solaris. It covers how Oracle Solaris 11 is being deployed in cloud computing and the unique optimizations in Oracle Solaris 11 for the Oracle stack. The session also offers a sneak peek at the latest technology under development for Oracle Solaris and what customers can expect to see in the next major release. The speaker is joined by a key customer executive who shares the benefits and value experienced with Oracle Solaris and the business challenges solved.
General Session
SPARC: Software in Silicon for Security and Speed. New SPARC design is probably M7.
General Session: Software in Silicon and SPARC Outlook—Secure, Smarter Database/Applications [GEN6421]
Masood Heydari, SVP, Hardware Development, Oracle
Software-in-silicon technology is the most important development in enterprise computing of the decade, created by Oracle’s unique vision of breakthrough microprocessor and server designs through coengineering with the database, applications, and Java. This session will help you understand the advanced features that secure your application data with less effort, accelerate the performance of the Oracle Database In-Memory option of Oracle Database 12c, and run Java middleware in the most efficient and fastest-possible way. You will also understand Oracle’s strategy for creating the SPARC server technology that is changing the way customers look at their cloud infrastructure and IT system and will hear about the outlook for systems based on the new SPARC design Oracle has just unveiled.
General Session
The SPARC M7 appears to be a soon to be realized reality, with various discussions coming up in OpenWorld and Java One!

Conclusions:

Firmware being released on SPARC Servers are a clear indication of what is here. Oracle will be discussing the usage of SPARC M7 as a reality with live labs. If you are building Network Management platforms, this is the time to start your planning for hardware acquisition, to get the most "bang for the buck".



 


Friday, August 21, 2015

Coming Soon: OracleWorld 2015 for Remote Management

[San Francisco California, courtesy Oracle Corporation]

Coming Soon: OpenWorld 2015 (for Remote Management)

The place to be will be Oracle OpenWorld in San Francisco, California during October 25-29.OpenWorld will offer many seminars to attend, but there is also JavaOne! Register & Fly to one of the most anticipated conferences of 2015. Why is this so anticipated?

Seminars & Sessions

There are a variety of seminars, conferences, and hands-on lab activities for just about everything imaginable. For people who are involved in Network, Systems, Database, and Application Management - the most applicable items are listed below. If you find some of these interesting, then you don't want to miss it!

For the Network, Operating System, Database, and Application Management staff:
Way Beyond the Basics: Oracle Enterprise Manager Monitoring Best Practices [CON9721]
Ana Maria Mccollum, Director of Product Management, Oracle
Oracle Enterprise Manager monitoring plays a critical role in enabling IT to provide highly available services to its business. As enterprises optimize their monitoring processes and evolve to adopt Oracle Cloud Platform services, Oracle Enterprise Manager continues to meet these new requirements with significant planned enhancements in monitoring. These include enhancements to metric extensions, adaptive thresholds, user-defined target properties, and corrective actions as well as new capabilities such as the incident manager dashboard, auto event grouping, the export/import of rule sets, and brownouts for planned/unplanned outages. In addition to new features, this session reviews best practices for implementing an effective and scalable monitoring solution.
Conference Session
Oracle Enterprise Manager: One Manager to Rule Them All - Ops Center and Oracle VM Manager.
Upgrading Oracle Enterprise Manager: Why and How [CON9729]
Akanksha Sheoran Kaler, Principal Product Manager, Oracle
The upcoming release of Oracle Enterprise Manager brings, for the first time, a converged management for Oracle hardware and software. It also includes exciting new enhancements to uninterrupted monitoring, hybrid cloud management, and engineered systems management that make the upgrade worth it. This session outlines the new capabilities and provides best practices to seamlessly upgrade your existing environment to the upcoming release of Oracle Enterprise Manager.
Conference Session
Do you use Oracle Middleware? Oracle offers Middleware as a Service with Oracle Enterprise Manager.
Realizing MWaaS on the Private Cloud Using Oracle Enterprise Manager [CON4627]
David Nims, UNIX Platform Architect, Fiserv
Umesh Panwar, Sr. Platform Engineer, Fiserv
Wojciech Serafin, Oracle
Provisioning a large number of middleware assets across multiple environments in a homogeneous way could be a challenging task for the IT organization. Enterprises are looking for an automated process for provisioning these assets within their data center and managing all application lifecycle management tasks using prebuilt flows. This session discusses how Fiserv, a worldwide provider of financial services technology, leveraged Oracle Enterprise Manager to provision and manage more than 500 middleware domains. The solution provided enterprise capabilities for automated, fast, simple, flexible, and reliable deployments based on the Oracle Enterprise Deployment Guide and significantly reduced time to market to provision new applications on its private cloud.
Conference Session
Security for Network Management on your mind? Oracle Solaris 11 with Security Auditing Framework.
Assessing, Reporting, and Customizing the Security Compliance in Oracle Solaris 11.2 [HOL4645]
Qianqian Chen, Oracle
Ling-yun Li, Principal Software Engineer, Oracle
Richard Liu, Senior Software Engineer, Oracle
Report Compliance is one of the new security features introduced in Oracle Solaris 11.2 that provides a framework for assessing and reporting the compliance of an Oracle Solaris system to a given security benchmark. In this lab, learn how to install the Report Compliance tool, run an assessment on the hosting Oracle Solaris against the Oracle Solaris baseline benchmark, generate an HTML report for review, and rerun the assessment after a quick remediation. And last, learn how to customize the benchmark by adding a user-defined check. After this lab, you will understand what Report Compliance is and how to use it to audit the security compliance of an Oracle Solaris system, and have an overall idea of how to customize a benchmark in case of need.
HOL (Hands-on Lab) Session
Use Java for Network & Systems Management (NSM)? Optimize using Solaris 11 with DTrace.
Uncover JDK 8 Secrets Using DTrace on Oracle Solaris 11 [HOL6427]
Gary Wang, Manager, Oracle
Yu Wang, Software Engineer, Oracle
Xiao-song Zhu, Principal Software Engineer, Oracle
JDK 8 is the most innovative version of Java ever. It brings many new features to the Java platform, such as Lambda Expressions, Streams, and Functional Interfaces. For the programmers, these features are easy to use; however, it is hard to understand their internal mechanisms. In this lab, learn how to use the Oracle Solaris 11 DTrace feature to find out how a Java Virtual Machine (JVM) implements these new features, including Streams Pipeline, Lambda Parallelism, Lazy Evaluation, and Recursion Optimizing.
HOL (Hands-on Lab) Session
How big can you make your Java based NSM? Scaling success stories by other applications.
Scaling to 1,000,000 Concurrent Users on the JVM [CON7220]
Jo Voordeckers, Senior Software Engineer, Livefyre
Livefyre built a platform that powers real-time comments and curated social media for some of the largest websites, such as CNN, Fox, Sky, CBS, Coca-Cola, HBO, CNET, Universal Music Group, and Break. On average it deals with one million concurrent users on its systems. Java EE will get you a long way, but with these numbers, the company needed to resort to some often-overlooked computer science tricks and techniques to scale its microservices architecture to handle as many as 100,000 active concurrent connections per JVM. This session covers some of the data structures, patterns, best practices, and datastores that Livefyre uses to make this all happen and keep it running. If you’re in a company with growing scalability pains, this session is for you.
Conference Session
Got Engineered Systems? Take care of them.
Monitor Engineered Systems from a Single Pane of Glass: Oracle Enterprise Manager 12c [UGF10288]
Alfredo Krieg, Sr. Oracle Enterprise Cloud Administrator, The Sherwin Williams Company
Oracle Enterprise Manger 12c provides comprehensive and centralized monitoring capabilities for Oracle engineered systems including Oracle Exadata Database Machine, Oracle Exalogic Elastic Cloud, and Oracle Exalytics In-Memory Machine. This presentation outlines the steps required to discover and monitor Oracle engineered systems, as well as the challenges faced and the benefits of using Oracle Enterprise Manager Cloud Control 12c to provide Oracle Exadata health-check reports.
User Group Forum Session
Got Databases? Best practices for managing hundreds or thousands.
How to Upgrade Hundreds or Thousands of Databases in a Reasonable Amount of Time [CON8375]
Mike Dietrich, Master Product Manager, Oracle
Roy Swonger, Sr Director, Software Development, Oracle
Many customers now have database environments numbering in the hundreds or even thousands. This session addresses the challenge of maintaining technical currency of such an environment while also containing upgrade and migration costs at a reasonable level. Learn from Oracle Database upgrade experts about product features, options, tools, techniques, and services that can help you maintain control of your database environment. You will also see examples of how real customers are successfully meeting this challenge today.
Conference Session
Network & Systems Management: The Future of Oracle Enterprise Manager.
The Future of Oracle Enterprise Manager: What’s Next? [CON9708]
Sudip Datta, Vice President of Product Management, Oracle
Oracle has been hard at work on the next major release of Oracle Enterprise Manager, and in this session the speakers are excited to give you a sneak preview of what’s coming. Learn about top new features including Integrated Hardware Management and Federated Enterprise Manager, upcoming Oracle Cloud integrations, and improvements across private cloud and stack management capabilities. Join this session for a glimpse of the future of Oracle’s on-premises private and hybrid cloud management capabilities.
Conference Session
Network & Systems Management: Best Practices with High Availability for Oracle Enterprise Manager
Practical Tips for Oracle Enterprise Manager High Availability and Diagnostics [CON9726]
Angeline Dhanarani, Senior Product Manager, Oracle
Many data centers have come to rely on Oracle Enterprise Manager as their management tool for mission-critical Oracle infrastructure and applications. As such, it is critical to ensure that the Oracle Enterprise Manager deployment is highly available and secure and performs optimally. This session shares best practices for managing an Oracle Enterprise Manager deployment with reduced effort while still ensuring that objectives are met. Strategies include configuring highly available deployments, migrating to a replication-based disaster recovery solution, deploying across networks using a new multiproxy server feature, monitoring and diagnostics of critical subsystems, and securing the Oracle Enterprise Manager infrastructure.
Conference Session
Network & Systems Management: From hundreds to hundreds of thousands of assets using OEM.
Scaling the Limits of the Cloud with the New Oracle Enterprise Manager [CON9731]
Mithun Shankar, Senior Principal Product Manager, Oracle
The cloud is BIG. The cloud is VAST. With cloud computing, the scale of IT has changed from a few hundred assets to hundreds of thousands of assets. This necessitates a newly engineered monitoring, automation, and reporting framework that is nimble, scalable, and real time. The new release of Oracle Enterprise Manager introduces real-time monitoring, a scalable job system that can integrate with the industry’s leading automation frameworks like Chef, and a comprehensive reporting infrastructure. This session covers these new enhancements along with anecdotal experiences from Oracle’s own cloud operations.
Conference Session
Building your own embedded Network Management probe in Java?
Alexander Belokrylov, Principal Product Manager, Oracle
This how-to session demonstrates a develop/build/deploy/debug/execute cycle set from scratch. It is based on Java ME Embedded in combination with various boards, such as the Raspberry Pi, the Freescale K64, and a device emulator. The session includes guidelines and tips on installing the Java ME Embedded SDK and runtime. It also touches on the key aspects of application development and troubleshooting in a simple demo that involves the basic concepts of working with various I/O devices.
Tutorial
Oracle Enterprise Manager: Provisioning Databases, without DBA'
Gustavo Rene Antunez, DBA Team Lead, Pythian
With the newest version of Oracle Database 12c and its multitenant option, we are moving toward an era of provisioning databases to our clients faster than we ever could, even leaving out the DBA and allowing the developers and project leads to provision themselves the database that they need. This presentation guides you through the different ways you can provision data from one Oracle Database to another using Oracle Enterprise Manager 12c.
User Group Forum Session
Oracle Enterprise Manager: Database as a Service
PDBaaS: Oracle Database 12c, Multitenant, and Oracle Enterprise Manager 12c [CON4628]
Krishna Kapa, Pepsico
Malla Santosh, Manager, GCSI, EM Foundation, Oracle
Database as a service (DBaaS) offers organizations accelerated deployment, elastic capacity, greater consolidation efficiency, higher availability, and lower overall operational cost and complexity. Oracle Database 12c provides an innovative multitenant architecture featuring pluggable databases that make it easy to offer DBaaS and consolidate databases in the clouds. This session showcases the implementation of pluggable database as a service (PDBaaS) using Oracle Enterprise Manager 12c and the benefits of consolidating databases into the Oracle Database 12c multitenant architecture, rapid provisioning of pluggable databases using the self-service portal, and maintenance of the pluggable databases.
Conference Session
Oracle Enterprise Manager OpsCenter, My Oracle Support, Maintenance, Upgrades, Patching.
Best Practices for Oracle Solaris Maintenance and Upgrades [CON8705]
David Buxton, Principal Technical Support Engineer, Oracle
Raphy Pallikunnath, Manager, Solaris & Network Technology Service Center, Oracle
Unleash the potential of Oracle Solaris, with an insight into technical resources and proactive support tools. This session looks at best practices for maintaining and upgrading Oracle Solaris. See how to leverage the My Oracle Support portal for both reactive and proactive issues, along with how the My Oracle Support Community can put you in touch with a wealth of knowledge. Enjoy a voyage of discovery to see the benefits and possibilities of using Oracle Enterprise Manager Ops Center features to help maintain and support your data center assets.
Conference Session
Oracle Enterprise Manager: Managing MySQL
Oracle Enterprise Manager for MySQL Database—Latest Features [CON4507]
Carlos Proal Aguilar, Senior Software Developer, Oracle
This session provides an overview of how to use the latest Oracle Enterprise Manager plugin release for MySQL Database to monitor database connections, index usage, replication status, compliance scores, and other key configuration and performance metrics.
Conference Session
Oracle Enterprise Manager: Hands-On for MySQL
Practical Overview of the Latest Features of Oracle Enterprise Manager for MySQL [HOL4522]
Carlos Proal Aguilar, Senior Software Developer, Oracle
In this hands-on lab, participants install and use the Oracle Enterprise Manager plugin release for MySQL to monitor database connections, index usage, replication status, compliance scores, and other key configuration and performance metrics.
HOL (Hands-on Lab) Session
Oracle Enterprise Manager: Monitoring Exadata
Oracle Exadata Monitoring and Management Best Practices [CON9727]
Ashish Agrawal, Group Product Manager, Oracle
Oracle Enterprise Manager uses a holistic approach to manage Oracle Exadata Database Machine, providing comprehensive performance and lifecycle management from testing and deployment to proactive monitoring and ongoing maintenance across the entire engineered system. In this session, hear about new capabilities in the upcoming Oracle Enterprise Manager plugin for Oracle Exadata that includes Oracle Exadata virtualization provisioning and monitoring, the ability to monitor the latest Oracle Exadata hardware, Exacheck (ORAchk), Exadata Sparse Cloning, and the Automatic Service Request capability for Oracle Exadata hardware targets in Oracle Enterprise Manager.
Conference Session
Software Defined Networks with Security and The Cloud
Network and Security Function with Oracle SDN Virtual Network Services [HOL10372]
With cloud data center architectures requiring agility and flexibility in deploying on-demand network services, the traditional approach with purpose-built physical network appliances does not meet the requirements. The Virtual Network Services feature of Oracle SDN (Software Defined Networking) provides the ability to deploy on-demand network services such as firewall, router, load balancer, virtual private network (VPN), and network address translation (NAT) in a single virtual appliance, eliminating the need for proprietary fixed function devices. With centralized management, secure multitenancy, and on-demand provisioning, this allows cloud-enabled data centers to be agile and elastic.
HOL (Hands-on Lab) Session
Monitoring Oracle Exadata Platforms
Get Under the Hood with Oracle Exadata Monitoring [CON10169]
Farouk Abushaban, Senior Principal Technical Analyst, Oracle
In this session, learn how to quickly set up complete monitoring for your Oracle Exadata Database Machine. Our subject matter expert and global technical lead in Oracle Enterprise Manager and Oracle Exadata support shares knowledge gained from working with customer deployments worldwide. Specific topics covered include common challenges, best practices, and new features to get your complete Oracle Exadata Database Machine stack monitored using Oracle Enterprise Manager Cloud Control.
Conference Session
Oracle Enterprise Manager Demonstrations and Features
Oracle Enterprise Manager: The Complete Solution and Oracle’s Best-Kept Secrets [CON9715]
Amit Ganesh, Vice President Enterprise Manager, Oracle
Come to this informative session to learn about the breadth of capabilities in Oracle Enterprise Manager from the head of Oracle Enterprise Manager Product Development. Through a series of product demonstrations, you’ll gain exposure to some of the most powerful features in the Oracle Enterprise Manager product family and see how they work together as a solution. Learn how to get the most out of the features you already use every day as well as how to leverage features you might not yet know about. Finally, receive expert guidance on how to maximize your Oracle OpenWorld experience to understand all that Oracle Enterprise Manager has to offer.
Conference Session
Maintaining and Supporting Oracle Enterprise Manager
Best Practices for Maintaining and Supporting Oracle Enterprise Manager [CON8671]
Rachel Bridden, Principal Technical Support Engineer, Oracle
Marilyn Roncati, Director of Lifecycle Management, Oracle
In this session, learn about best practices, tips, and tools for maintaining and getting the most out of Oracle Enterprise Manager. Experts from Oracle Support offer knowledge gained from working with Oracle customers worldwide. They look at patching, upgrades, issue resolution, and more. Specific topics covered include Oracle Enterprise Manager metrics and health checks, remote diagnostics, communities, and how to receive priority service request handling.
Conference Session
Managing Security with Oracle Enterprise Manager
Raising the Ante on Security with Oracle Enterprise Manager [CON9719]
Angeline Dhanarani, Senior Product Manager, Oracle
Ana Maria Mccollum, Director of Product Management, Oracle
In today’s highly connected world, security is a critical area of concern for both IT and the business. Security teams demand compliance with security best practices and corporate security standards. In this session, learn practical strategies to help you adhere to these standards using Oracle Enterprise Manager. Topics include external authentication, authorization, user management, public and private roles, managing passwords, privilege delegation providers, and secure communications, with a focus on key upcoming Oracle Enterprise Manager enhancements, such as managing DBSNMP accounts, new privileges to manage users, new fine-grained database access privileges, and Transport Layer Security-based communications between Oracle Enterprise Manager components.
Conference Session
Automating Solaris Management with Puppet
Automating Oracle Solaris Administration with Puppet [HOL10359]
Geoffrey Gardella, Senior Quality Assurance Engineer, Oracle
Cindy Swearingen, Senior Product Manager, Oracle
Oracle Solaris 11 integrates Puppet, a configuration management solution that you can use to automate Oracle Solaris administration tasks. In this lab session, learn how to use Puppet to automate Oracle Solaris lifecycle management tasks.
HOL (Hands-on Lab) Session

Thursday, August 20, 2015

M7: Next Generation SPARC and Next Generation Data Center

[Oracle SPARC M7 Die, Courtesy The Register]

Abstract:

The SPARC Processor family from Sun Microsystems had existed for nearly 30 years. SPARC was an early contender in the 64 bit processing market while most commodity processors were still 32 bits or fewer. With the purchase of Sun Microsystems by Oracle, SPARC development continued and produced the fastest processor on the planet. Oracle promised a day where Oracle would have one processor for both the T and M platforms. The latest generation of processor finally appears able to unify the T and M system lines. In December 2014, the Hot Chips 26 Symposium in 2014 conference material was released to the general public, illustrating what is coming in 2015. Happy New Year!

[M7 32x Socket Interconnect, courtesy Enterprise Tech]

SPARC M7 at Hot Chips 26 in 2014:

Oracle announced during the Hot Chips conference that the new M7 processor and system would be released in 2015. Some notable reviews of the Oracle SPARC M7 included:
  • [2014-08-13] Timothy Pricket Morgan from Enterprise Tech Systems Edition
    Oracle Cranks Up The Cores To 32 With Sparc M7 Chip
    When asked about what the performance advantage would be comparing an InfiniBand or Ethernet cluster running Oracle RAC and the Sparc M7-Bixby setup using the non-coherent memory clusters, Fowler said that the difference “would not be subtle.”
  • [2014-08-18] Simon Sharwood from The Register
    Oracle reveals 32-core, 10 BEEELLION-transistor SPARC M7
    New chip scales to 1024 cores, 8192 threads 64 TB RAM, at speeds over 3.6GHz
  • [2014-08-12]  James Niccolai  from PC World
    Oracle's Sparc M7 chip to supercharge in-memory computing
These reviews were helpful in understanding the market's take on the hardware announcement from Oracle, but the actual presentation was released in December 2014 had the details.

[Oracle 2010 SPARC Road Map, courtesy Enterprise Tech]

Hot Chips Oracle Presentation:

Stephen Phillips, the Senior Directory SPARC Architecture from Oracle, gave the presentation. He was involved in the architecture for T2+ (Victoria Falls) and later T-Series; M5, M6, M7 Processors, placing him in the cross-hairs of the delivery of the 2010 roadmap above. The Hot Chips 26 - August 12, 2014 (Big Iron Presentations) was included in Video (Video Start-End: 32:26-1:04:00) as well as High-Resolution PDF.

It should be noted, that the clustering of the M7 produces a system capable of remote memory sharing with up to 64 sockets, make the formerly released 5 year road map (pictured above) astounding accurate, illustrating the intense fidelity that Oracle offered to the SPARC and UNIX communities.

[M7 Decompression and Query Offload Engine, courtesy EnterpriseTech]

SPARC M7 In-Silicon Enhancements:

As one watches the presentation and reviews the high resolution slides, the following is a short set of notes, highlighting what some would consider key aspects.

Slide 3 - Recent history of SPARC
Slide 5 - M5 CPU - 20nm process; 32 Cores per Socket; 4 clustered cores; S4 enhanced core
Slide 6 - S4 Core - Dynamic Threading (1-8 threads) for speed & throughput; Faster Live migration
Slide 7 - Core Cluster - 1.5x Larger L2 Cache; 2x Greater Core Bandwidth
Slide 8 - Level 3 Cache & Network: 2.5x-5x Bandwidth; 25% Less Latency; HW Accelerator Access
Slide 9 - OVM "Aware" Solaris Process Groups by Core Cluster &, L3 Cache Partition
Slide 10 - Power aware in silicon; auto-adjusts voltage & frequency according to policy
Slide 11 - 2x-3x Memory Bandwidth; Live DIMM Retirement; Memory Lane Failover;
Greater than 2x PCIe performance
Slide 12 - Performance increase ~3x over M6
Slide 13 - Live Production Data Integrity Checking (for buffer overrun protection)
Slide 14 - Fine Grain Memory Migration for JAVA (for concurrent  operations of middleware)
Slide 15 - Virtual Memory Masking for Java Runtime (embed object state into unused 64 bit)
Slide 16 - Decompression & Query accelerators for Oracle 12c (row & column for OLTP & OLAP)
Slide 17 - 8x Fused Decompression + Query Accelerators
Slide 18 - High performance for in-memory database without OS intervention
Slide 19 - 10-to-1 Decompression improvement of 1x query pipeline to 1x T5 (S3) Thread
Slide 20 - Third Party benefit through tool-chain
Slide 21 - Glue-less 1-8 socket support, like the T5
Slide 22 - SMP Scalability comprising 32 M7 SPARC Processors; 4 socket physical domains
Slide 23 - Reliable & Secure Shared Memory Clustering; 64 M7 sockets in a cluster; >1 failure tolerance
Slide 24 - Coherent Memory Cluster comprising 64 M7 SPARC Processors; secure foreign memory

The use of hardware application accelerators has proven to be a massive game changer in the industry for Oracle, as SPARC continues where others have failed.

Conclusions:

In the world of Network Management, this particular hardware solution  enables massive network management scalability, the fastest virtualization migration technology, and provide the most reliable underlying infrastructure. Underpinning the SPARC processor technology includes 2x-10x "everything" over commodity Intel platforms, lashed together by various vendors. This is where high performance data centers which need a small footprint, low power utilization, enterprise software, and massive scalability will go.

Tuesday, August 18, 2015

ZFS: Flash & Cache 2015q1

ZFS: Flash & Cache 2015q1

Abstract:

The concept of Storage Tiering existed from the time computing came into existence. ZFS was one of the first mainstream file systems to think about automatic storage tiering during it's initial design phase. Advances in ZFS had been made to make better use of cache during recent times.

Multiple kinds of Flash?

Flash comes primarily in two different types: highly reliable single-level cell (SLC) memory and multi-level cell (MLC) memory. The EE Times published a technical article describing them.
SLC... NAND flash cells... Both writing and erasing are done gradually to avoid over-stressing, which can degrade the lifetime of the cell.  
MLC... packing more than one bit in a single flash storage cell... allows for a doubling or tripling of the data density with just a small increase in the cost and size of the overall silicon. 
The read bandwidths between SLC and MLC are comparable
If MLC packs so much more data, why bother with SLC? There is no "free lunch", there are differences between SLC and MLC in real world applications, as the IEEE article describes.
MLC can more than double the density [over SLC] with almost no die size penalty, and hence no manufacturing cost penalty beyond possibly yield loss.
...
Access and programming times [for MLC] are two to three times slower than for the single-level [SLC] design.
...
The endurance of SLC NAND flash is 10 to 30 times more than MLC NAND flash
...
difference in operating temperature, are the main reasons why SLC NAND flash is considered industrial-grade
...
The error rate for MLC NAND flash is 10 to 100 times worse than that of SLC NAND flash and degrades more rapidly with increasing program/erase cycles
...
The floating gates can lose electrons at a very slow rate, on the order of an electron every week to every month. With the various values in multi-level cells only differentiated by 10s to 100s of electrons, however, this can lead to data retention times that are measured in months, rather than years. This is one of the reasons for the large difference between SLC and MLC data retention and endurance. Leakage is also increased by higher temperatures, which is why MLC NAND flash is generally only appropriate for commercial temperature range applications.
It is important to understand the capabilities of Flash Technology to determine how to gain the best economics from the technology.

ZFS Usage of Flash and Cache

The usage of MLC Cache in a proper storage hierarchy is impossible to omit. The doubling of storage capacity at almost no cost impact is a deal nearly too great to ignore! How does one place such a technology into a storage system?

When a missing block of data can result in loss of data on the persistent storage pool, then a highly reliable Flash is required. The ZFS Intent Log (ZIL), normally stored on the same drives as the managed data set, was architected with an external Syncronous Write Log (SLOG) option to leverage SLC NAND Flash. The SLC flash units are normally mirrored and placed in front of all writes going to the disk units. There is a dramatic speed improvement whenever writes are committed to the flash since committing the writes to disk take vastly longer, and those writes can be streamed to disk after random writes were coalesced to Flash. This was affectionately referred to as "LogZilla".

If the data is residing on persistent storage (i.e. disks), then the loss of a block of data merely results in a cache miss, so the data is never lost. With ZFS, the Level 2 Adaptive Read Cache (L2ARC) was architected to leverage MLC NAND Flash. There is a dramatic speed improvement whenever reads hit the MLC before going to disk. This was affectionately referred to as "ReadZilla".

Two things to be cautious about, regarding flash... electrons disappear over time and just reading data can cause corruption of data. To compensate for factors such as these, ZFS was architected with error detection & correction, inherently in the file system.

Performance Boosts in ZFS from 2010 to 2015

ZFS has been running in production for a very long time. Many improvements have been made recently, in order to improve on "State of The Art" of Flash and Disk!

Re-Architecture of ZFS Adaptive Read Cache

Consolidate Data and Metadata Lists
"the reARC project.. No more separation of data and metadata and no more special protection. This improvement led to fewer lists to manage and simpler code, such as shorter lock hold times for eviction"
Deduplication of ARC Memory Blocks
"Multiple clones of the same data share the same buffers for read accesses and new copies are only created for a write access. It has not escaped our notice that this N-way pairing has immense consequences for virtualization technologies. As VMs are used, the in-memory caches that are used to manage multiple VMs no longer need to inflate, allowing the space savings to be used to cache other data. This improvement allows Oracle to boast the amazing technology demonstration of booting 16,000 VMs simultaneously."
Increase Scalability through Diversifying Lock Type and Increasing Lock Quantity
"The entire MRU/MFU list insert and eviction processes have been redesigned. One of the main functions of the ARC is to keep track of accesses, such that most recently used data is moved to the head of the list and the least recently used buffers make their way towards the tail, and are eventually evicted. The new design allows for eviction to be performed using a separate set of locks from the set that is used for insertion. Thus, delivering greater scalability.
...
the main hash table was modified to use more locks placed on separate cache lines improving the scalability of the ARC operations"
Stability of ARC Size: Suppress Growths, Smaller Shrinks
"The new model grows the ARC less aggressively when approaching memory pressure and instead recycles buffers earlier on. This recycling leads to a steadier ARC size and fewer disruptive shrink cycles... the amount by which we do shrink each time is reduced to make it less of a stress for each shrink cycle."
 Faster Sequential Resilvering of Full Large Capacity Disk Rebuilds
"We split the algorithm in two phases. The populating phase and the iterating phase. The populating phase is mostly unchanged... except... instead of issuing the small random IOPS, we generate a new on disk log of them. After having iterated... we now can sort these blocks by physical disk offset and issue the I/O in ascending order. "
On-Disk ZFS Intent Log Optimization under Heavy Loads
"...thundering herds, a source of system inefficiency... Thanks to the ZIL train project, we now have the ability to break down convoys into smaller units and dispatch them into smaller ZIL level transactions which are then pipelined through the entire data center.

With logbias set to throughput, the new code is attempting to group ZIL transactions in sets of approximately 40 operations which is a compromise between efficient use of ZIL and reduction of the convoy effect. For other types of synchronous operations we group them into sets representing about ~32K of data to sync."
ZFS Input/Output Priority Inversion
"prefetching I/Os... was handled... at a lower priority operation than... a regular read... Before reARC, the behavior was that after an I/O prefetch was issued, a subsequent read of the data that arrived while the I/O prefetch was still pending, would block waiting on the low priority I/O prefetch completion. In the end, the reARC project and subsequent I/O restructuring changes, put us on the right path regarding this particular quirkiness. Fixing the I/O priority inversion..."
While all of these improvements provide for a vastly superior file system, as far as performance is concerned, there is yet another movement in the industry which really changed the way flash is used in Solaris with ZFS. As flash becomes less expensive, it's use will increase in systems. A laser was placed upon optimizing the L2ARC, making it vastly more usable.

ZFS Level 2 Adaptive Read Cache (L2ARC) Memory Footprint Reduction
"buffers were tracked in the L2ARC (the SSD based secondary ARC) using the same structure that was used by the main primary ARC. This represented about 170 bytes of memory per buffer. The reARC project was able to cut down this amount by more than 2X to a bare minimum that now only requires about 80 bytes of metadata per L2 buffers."

ZFS Level 2 Adaptive Read Cache (L2ARC) Persistence on Reboot
"the new L2ARC has an on-disk format that allows it to be reconstructed when a pool is imported... this L2ARC import is done asynchronously with respect to the pool import and is designed to not slow down pool import or concurrent workloads. Finally that initial L2ARC import mechanism was made scalable with many import threads per L2ARC device."
With large storage systems, regular reboots are devastating to the performance of the cache. The process of flushing the cache and re-populating them will shorten the life span of the flash. With disk blocks already existing in L2ARC, performance  improve. This also brings the benefit of inexpensive flash media as persistent storage, while competing systems must use expensive Enterprise Flash in order to facilitate persistent storage.

Conclusions:

Solaris continues  to advance using engineering and technology to provide higher performance at a lower price point than competing solutions. The changes to Solaris continues to drive down the cost of high performance systems at a faster pace than mere dropping in price of commodity hardware that competing systems depend upon.