Future of Storage: Flash

Future of Storage: Flash

Abstract

With ever increasing storage size increases, cost decreases, and performance increases - it seems like Flash storage will soon be a winner in the Managed Services arena. Others are writing about this technology, such as StorageMojo and The Register. Understanding where the technology is going is a good start, but understanding your technological bottlenecks is required for application.

What's New?

The flash DIMM format is an opportunity to significantly change the computing industry for the long term. Pictured below is a Sun Flash DIMM, in the form factor of common laptop memory.


Since hard disk drives are normally always spinning and generating massive quantities of heat - failure rates were high. High failure rates necessitated easy access in arrays through front and rear chassis access. Removal of the heat generation and mechanical movement increases reliability - so why would one need to suck up space in the front or rear of a rack with the center of the rack mount unit being mostly empty?

DIMM is the way to go.

What's Up and Coming?

Sun Systems designer, Andy von Bechtolsheim discussed the use of flash a quarter ago, both in the regular market, as well as in the future of the Sun marketing products.

One example of a core building block was a 4 Terabyte 1U high storage unit.


While 4 Terabytes in a 1U high rack space may not make people jump for joy, the news is really around the benefits for a dramatic increase in performance. When one can increase performance 100x, use 1/100th the rack space, and use orders of magnitude less power - this can drive change in any business.

Sun briefly posted a PDF of the F5100 storage platform, but this document was pulled. Google still has the HTML version of the "Sun Storage F5100 Flash Array Getting Started Guide" document, although it is fairly stripped of visual content and structure. You can see from the HTML that it was published in July of 2009.

What's Here Today

What is very comforting is that this up & coming technology is already supported by standard storage management tools - Sun StorageTek Common Array Manager (CAM).


The F5100 Flash Array is steadily appearing in more standard Sun documentation, for example the "Sun StorageTek Common Array Manager User Guide for Open Systems". There is also a PDF of this guide (as well as others) available, for common consumption.

Application in Network Management

For Open Source polling & graphing software, the I/O becomes inhibitive in large installations. The need for many spindles in order to keep up with the data read & write rates had created the architectural need to split the database from the polling software. With massive quantities of data coming into a database on very rapid & regular polling rates - the need for archiving this data becomes increasingly important, but secondary to the performance since massive numbers of spindles will leave high capacity drives mostly empty.


What would happen if the artificial need for breaking storage away from the pollers disappeared?

The architecture could simplify and re-consolidate onto a single server with multiple [virtual] pollers. The result would be a reduced level of system complexity (fewer servers, switch ports, physical ports, drive interconnects, etc.), increase performance (eliminate the need for massive external storage), increase reliability (fewer moving mechanical parts), and overall decrease costs.

Microsoft IIS Vulnerabilities Across Releases

Microsoft IIS Vulnerabilities Across Releases

New IIS attacks (greatly) expand number of vulnerable servers

The Register published a short article of concern for those of us in the Network Management industry, where we customer or internet facing platforms for reporting delivery.

Microsoft continues to say that IIS5 running on Windows 2000 appears to be the only version that is vulnerable to attacks that can remotely execute malicious code on an underlying server. But it's now clear that hackers can target every version of IIS to cause denial-of-service attacks.

If you have a current or legacy IIS server - this may place your installation at risk. This is an piece of old code, meaning that historical code that you have not touched for awhile will be at risk. The risk centers around industry standard FTP protocol, one of the backbone protocols of the internet.

If Microsoft is not releasing patches for your old release of IIS, time to think about replacing that old portal.

Microsoft rejects call to fix SQL password-exposure risk

Microsoft rejects call to fix SQL password-exposure risk



Abstract

Most serious Managed Services Element Management Platforms, which depend on external databases, traditionally do not depend on databases such as Microsoft SQL. This article illustrates one of the reasons: security.

The Problem

"Applications go to great lengths to obfuscate passwords when they are needed within the software, and should not store passwords as 'clear text,' either in memory (as is the case with this vulnerability) or on disk," Sentrigo's advisory stated.

Microsoft has rejected the company's calls to change the way the software handles passwords, saying people with administrative rights already have complete control of the system anyway.

The Response

"Microsoft has thoroughly investigated claims of vulnerabilities in SQL Server and found that these are not product vulnerabilities requiring Microsoft to issue a security update," a spokesman wrote in an email. "An attacker who has administrative rights already has complete control of the system and can install programs; view, change, or delete data; or create new accounts with full user rights."

What this means to Network Management

The problem with passwords being stored in the clear is not that an infected system could have data destroyed on it, but rather other systems what work with that infected system could be infected!

Of course, behaviors like this are rampant with Day-0 Exploits, Microsoft SQL Worms, Microsoft Windows Viruses, etc. Another place to get passwords by malware is just another reason not to implement such a system in an area where customer managed devices are routable.

If a system is storing passwords for thousands of managed systems in the clear, an infection of a central system could be disastrous for the managed customer edge devices.

A developer in a company may have the option to secure passwords or not - but if the developer in a company ever has to meet a PCI audit and the vendor does not offer that option, then the company providing the managed services is placed in tremendous risk.

Multi-Node Cluster Shared Nothing Storage

Multi-Node Cluster Shared Nothing Storage

Abstract

A number of months back, a new release of Sun Cluster was released, in conjunction with OpenSolaris 2009.06. This release offered a new architecture for a lower cost fail-over cluster capability using Shared-Nothing Storage. This paper discussed the benefits using a broader implementation plan to further reduce costs and increase scalability.

Shared Nothing Storage

With the advent of ZFS under Solaris and ComStar under OpenSolaris, there is a new no-cost architecture in the world of high-availability under Sun - Shared Nothing Storage.

The benefits are clear in this environment:

• External Storage is not required (with it's complexity and costs)


• Additional storage area network infrastructure is not required (with it's complexity and costs)


• The OS of the active node continually keeps all the local disks in sync (with virtually no complexity)

There are some drawbacks to this environment:

• Complete CPU capacity is needed on both platforms for peak CPU capacity for active applications.

Applications can be run under Node-1 while Node-2 is always kept up to date, ready for failover of the storage as well as the applications which are sitting on that storage pool.

Dual-Node Shared Nothing Storage

Some people may not bee too impressed - there is still a node which is completely unused. This additional node may be considered a pure cost in an H-A or D-R environment. This is not necessarily true, if other strategies are taken into consideration.

For example, a dual-active node, where individual internal storage could be leveraged on dual active nodes through dual initiators, to completely leverage CPU capacity on both nodes during peak times.

The benefits are clear in this environment:

• External Storage is not required (with it's complexity and costs)


• Additional storage area network infrastructure is not required (with it's complexity and costs)


• The OS of the active node continually keeps all the local disks in sync (with virtually no complexity)


• 200% CPU capacity on two platforms can be leveraged during peak usage times

There are some drawbacks to this environment:

• Fail-over of a single node results in reduction to 100% of CPU capacity

Applications can be run under Node-1 and Node-2 while disks on the opposing node is always kept up to date, ready for failover of the storage as well as the applications which are sitting on that storage pool.

Multi-Node Shared Nothing Storage

The dual-active node share nothing architecture seems very beneficial, but what can be done in very typical three-tier environments?

Considering how simple it is to move around pools as well as zones, multi-node clustering can be done with a couple of simple scripts.

For example, a triple-active node, where individual internal storage could be leveraged on all three active nodes through triple initiators, to completely leverage CPU capacity on all nodes during peak times.


The benefits are clear in this environment:

• External Storage is not required (with it's complexity and costs)


• Additional storage area network infrastructure is not required (with it's complexity and costs)


• The OS of the active node continually keeps all the local disks in sync (with virtually no complexity)


• 300% CPU capacity across all platforms can be leveraged during peak processing times


• Failover of a single node means only a decrease to 200% CPU processing capacity

Applications can be run under Node-1, Node-2, and Node-3 while disks on the opposing nodes are always kept up to date, ready for failover of the storage as well as the applications which are sitting on that storage pool.

Application in Network Management

What does this have to do with Network Management?

Very often, there are multiple platforms which are used on polling platforms, with a high-availability requirement on an embedded database. There is usually a separate cost for H-A kits for applications as well as databases.

Placing each of the tiers within a Solaris Container is the first step to business optimization, higher availability, and cost reduction.


As a reminder, Oracle RDBMS can legally be run within a CPU Capped Solaris 10 Container, in order to reduce CPU licensing costs, leaving plenty of CPU available for failing over applications from other tiers. As additional capacity is needed by the business, the additional license can be purchased and the cap extended to other cores on the existing platform.

Pushing down the H-A requirements to the OS level eliminates application & license complexities and enables drag-and-drop load balancing or disaster-recovery under Solaris 10 or OpenSolaris using Solaris Containers. Running a RDBMS within a Capped Solaris 10 Container gives the business the flexibility to buy/stage hardware without having to pay the unused cpu cycles until they are actually needed.

- - - - - - - - - - - - - - - - - - -

Update - 2009-01-07: Another blog posting about this feature:

Solaris tip of the week: iscsi failover with COMSTAR


Update - 2019-10-21: Previous "Solaris tip of the week" no longer exists, transferred post:
https://jaydanielsen.wordpress.com/2009/12/10/solaris-tip-of-the-week-iscsi-failover-with-comstar/

I've been researching HA iscsi configurations recently, and I'd like to capture and share what I've learned about the COMSTAR stack. I have a simple demo that you can use for your own experiments...

 

More Work With ZFS

More Work With ZFS

The Last Time...

The last time ZFS was covered, an description of overall features were covered. How to use all of those features was uncovered. This post will try to cover some of the other features.

ZFS Sharing Overview

ZFS centralizes all directory sharing into a single command structure and removes the needs to manage arcane configuration files to deal with issue such as configuration, status, and persistency.

ZFS Sharing Protocols

The new ZFS suite offers protocol sharing of iSCSI, NFS as well as SMB(CIFS). There is a catch to this: the ZFS host must support a kernel implementations of the protocols - SMB(CIFS) is only supported under more recent releases of OpenSolaris and iSCSI is only supported under the Solaris families.

ZFS Sharing Stopping and Starting

ZFS uses a property to determine whether a filesystem mount is going to be shared or not.

To stop NFS sharing of a ZFS filesystem.

servera/root$ zfs set sharenfs=off u201

To start an NFS sharing of a ZFS filesystem.

servera/root$ zfs set sharenfs=on u201

Sharing Status

On the same server, one can check the "share" command to see what is being shared from all protocols, persistent or not. A listing of domestic sharing protocols that can be checked are in a configuration file on the sharing host

servera/admin$ cat /etc/dfs/fstypes
nfs NFS Utilities
autofs AUTOFS Utilities
cachefs CACHEFS Utilities

servera/admin$ share
- /u000 anon=60001,rw=servera "" - /u201 rw ""

On a foreign server, one can check to see what is being shared (via NFS protocol), persistent or not. A listing of foreign protocols that can be checked are in a configuration file on the remote host.

serverb/admin$ cat /etc/dfs/fstypes nfs NFS Utilities autofs AUTOFS Utilities cachefs CACHEFS Utilities

serverb/admin$ dfshares servera RESOURCE SERVER ACCESS TRANSPORT servera:/cdunix servera - - servera:/u201 servera - -

For the share and dfshares command, if no protocol is specified, then the "nfs" protocol is the default. A ZFS filesystem shared over NFS can be done using the "share" and "dfshares" command.

Sharing and Persistence

In most historic POSIX systems, there is a file referred to as "sharetab" (or some derivative of it) to review the sharing of filesystems. This is effective against any underlying filesystem (i.e. UFS, VxFS, ZFS, etc.) In the example below, cdunix is not on a ZFS filesystem.

servera/admin$ cat /etc/dfs/sharetab /u000 - nfs rw /u201 - nfs rw

If one is running a pure ZFS environment, persistence is held as a property. You can see the status of the ZFS file share through a ZFS command.

servera/admin$ zfs get sharenfs u201
NAME PROPERTY VALUE SOURCE
u201 sharenfs on local

Checking all shared protocols through ZFS is also possible, through parsing "all" properties option.

servera/admin$ zfs get all grep share
u201 sharenfs on default
u201 shareiscsi off default
u201 sharesmb off default

Checking the share status for all protocols from a foreign server is not as elegant. Individual protocols must be used, such as the "dfshares" command.

A New Tablet on the Horizon?

A New Tablet on the Horizon?

Mac Tablet Rumors

There has been grumbling about Apple Macintosh "tablet" form factors for years, although the leaks have been getting more substantial as of late.

One such published rumor included a quote, from an enthusiast, which was highly criticized:
"The iTouch Tablet is about to change society as we know it."

This comment really all depends upon the execution by Apple.

Background to Successful Appliance Launches

What made the iPod, iPod Touch and iPhone successful?

Apple figured out how to change the industries in three markets: portable music players, PDA's, and Cell Phones. As a side note, Apple failed to create the PDA market with the Newton.

If Apple applies the same consideration into the Tablet market, they could do well.

Moving onto the Tablet

Making a successful consumer oriented appliance technology in a tablet which is easily manageable could be the key. Without the need for OS patches & upgrades in conjunction with virus updates, many traditional market outlets could choose the device over a portable PC or laptop:

• home theater entertainment
• eBook reading appliances
• libraries
• conference centers
• church pulpits
• university professors
• university students
• audio mixing consoles
• video special effects generators
• lighting control panels

Anyone who has used cheaply manufactured devices (made with little attention to software & hardware details) tire of the experience quickly (due to the issues inherent with trash design & manufacturing.) A large number of good ideas go to the technical graveyard when consumers believe they are poorly implemented the first time around. An Apple "iTablet" with the engineering of a Apple MacBook Air might be accepted rapidly.


Anyone who has used industrial devices, which have physical controls, spend a pretty penny for them - moving the technology to touch screen controls could result in a far better user experience and longevity in the device usage without substantial maintenance in cleaning sliders/pots.

If Apple does a tablet right, they could really revolutionize many industries.

Network Management

What does all of this have to do with Network Management?

• Network Management uses obtuse interfaces from a variety of vendors.
  
• The features from multiple vendors use significantly different interfaces, some of which provide poor user interface capabilities, and all are generally very expensive to implement on a per-user basis.
• People are becoming familiar with many "Web 2.0" features in every day life and these are not being backfilled into the Network Management arena by vendors.
• People are demanding more mobility and many Network Management vendors are not delivering these features by investing in a time of low revenue in an global economic recession.
  

How could these issues be resolved in a tablet?

• Remote Control capabilities (such as RDP, VNC, Telnet) are all available & widely distributed today in the iPod Touch and iPhone via Apple iTunes today, at a very reasonable cost (per user.)
• Remote debugging capabilities (such as Ping, Traceroute) are available & widely distributed today in the iPhone Touch and iPhone via Apple iTunes today, at a very reasonable cost (per user.)
• Corporations are already leveraging portables such as iPod Touch and iPhone via Microsoft Exchange support for corporate applications.
• VPN capabilities are built into remote devices like the iPhone today for remote capabilities.
  
• Using a standard interface, imposed by the iPod Touch or iPhone API's for multiple existing Web 2.0 applications, help users reduce barriers to entry through simplification, and would reduce training requirements for existing Network Management applications due to obtuse user interfaces by the vendors.
• Using a standard interface, imposed by the iPod Touch or iPhone API's for multiple existing Web 2.0 applications, would allow greater cross-vendor integration, since the platform would become the integration location.
• User interaction with Network Management maps and displays often use mouse clicks and drags, while API's in the iPod Touch and iPhone offer much more intuitive interactions such as dragging your finger or pinch.
• API integration of Google Maps into iPod Touch or iPhone with Location would reduce the burden of development on Network Management user interfaces - providing sophisticated geographical maps to the application provider without needing to create & license bloat-ware.
• API integration of Google Maps into iPod Touch or iPhone with Location would reduce the burden of use with Network Management user interfaces - providing a unified look-and-feel to what users expect in typical mapping applications while on foot or in their car
• People are already familiar with standard notification technology with their home computers through Web 2.0 interfaces and the mobile equivalents (through devices such as the Apple iPod Touch and Apple iPhone) - so the building of new notification technology filters by vendors becomes irrelevant, allowing users to use interfaces comfortable to them, while being productive more quickly when starting to use the standard interfaces
  
• The cost of a Apple "iTablet" hardware (or whatever it will be called) would be far less than the licensing per-user that is typically paid to a vendor on a per-seat charge, if standard interfaces could be developed to Web 2.0 environment.
• Not knowing what managed vendor equipment looks like is something that could be a thing of the past when leveraging software suite built into the Apple iPhone or iPod Touch - with the ability to have integrated photo library that sync's with a central database, a library could be kept up-to-date on all remote devices very easily, since the software management is built in.
  

Closing Thoughts

Will Apple release a tablet?

This is a great unknown, but all being said, moving Network Management applications to devices like the Apple iPod Touch and iPhone is a no-brainer.

A larger form factor in an appliance (that is not an Apple laptop) could easily infiltrate the Network Management business, especially if it can demonstrate cost savings during dire economic times.

Rock Cancellation Rumor, The Bizarre, and The Benefits

Rock Cancellation Rumor, The Bizarre, and The Benefits

Rock Cancellation Rumor

There has been much speculation concerning the rumor originating from the New York Times suggesting Sun Microsystems canceled the processor UltraSPARC RK called 'Rock'.

Understanding the history behind 'Rock' provides support to just about any rumor.

• UltraSPARC III+ was canceled by Sun
• UltraSPARC IV+ was released late by Sun, with excellent performance
  
• UltraSPARC V was canceled by Sun
• UltraSPARC T1 was deployed on-time by Sun, with excellent performance
• UltraSPARC T2 was deployed on-time by Sun, with excellent performance
• UltraSPARC T2+ was deployed on-time by Sun, with excellent scalability
• Sun Microsystems partnered with Fujitsu to release a joint server product line based around the SPARC64 processor
  
• UltraSPARC RK was delayed multiple times by Sun
  

Rock Unintersting?

There was very good historical information in many trade journals. One of the most bizarre quote regarding Sun Microsystems processor UltraSPARC RK or 'Rock' was from Dean McCarron, principal analyst at Mercury Research. His quote was widely reported in various trade magazines, such as:

• Sun's Rock Chip Doomed From The Start
• Sun's Rock doomed from start, analysts say
• Sun's Rock doomed from start, analysts say
  

"Even if Rock had made it to market, it would have been an uninteresting processor as companies like Intel and AMD are offering high-performance chips at more reasonable prices, said Dean McCarron, principal analyst at Mercury Research."

Suggesting 'Rock' is "uninteresting" demonstrates a level of ignorance beyond comprehension.

Rock Interested Audiences

'Rock' is an interesting processor to Computer Scientists since they have worked for decades trying to optimize single threaded applications. Thread bound applications run with very painful waits when they encounter cache misses on traditional proprietary (Intel & AMD) processors. These waits are a thing of the past on Rock with technologies such as:

• thread level parallelism
• thread level speculation
• transactional memory
• out-of-order retirement
• deferred queue

'Rock' is a very interesting processor in the academic world since companies like Intel and AMD have not recently pioneered computer science technologies in silicon to optimize single threaded applications. Implementations of theory in real silicon are very important to review academic thoughts and determine future implications. Implementations like 'Rock' are studied for decades.

'Rock' is a very interesting processor for business, military, and academia system performance of single thread bound applications - software threads on 'Rock' runs with very few of the painfully long waits on slow memory due to cache misses, commonly experienced with Intel and AMD processors. People who purchase systems expect their systems to be doing work, instead of sitting around idle.

'Rock' is a very interesting processor in the commercial world since accelerating legacy single thread-bound software allow for acceleration of existing software (which does not scale well with multiple threads) - something the major CPU developers (AMD and Intel) in rest of the market have been ignoring for a couple of years. If a single thread is the problem, newer CPU's from other vendors will not solve their performance problem, increase the thread bottlenecked performance, and increase the business profitability.

'Rock' is a very interesting processor in the investment community. Sun had pioneered the niche of multi-threaded hardware with the release of their 32 hardware thread UltraSPARC T1 processor - driving other vendors (Intel, AMD) to change their directions to start heavily threading their CPU's... discontinuing projects to speed existing single-threaded applications. The release of Rock would enable Sun to pioneer this lost niche, abandoned by the other vendors. Filling niches are very profitable to investors in those technologies.

'Rock' is very interesting to the environmentally conscience consumer. Very little work has been done recently in the market to increase the performance of single threaded software, with the exception of increasing clock rate, which drives up the costs to consumers in: hardware, cooling, and power consumption. Rock has been the exception - targeting increased single threaded performance without aggressively increasing clock rate and the negatives that go along with it.

'Rock' is a very interesting processor for enterprises struggling with consolidation efforts. Bundling 16 high-speed cores into a single chip which supports LDOM's at the firmware level and Solaris Containers at the OS Level provides a consolidation platform for legacy applications which are not highly-threaded and required high single threaded throughput.

Rock & Role in Network Management

Network Management infrastructure is greatly benefited by highly threaded underlying infrastructure. It is not unusual to tun run hundreds to thousands of polling threads on centralized network management platforms. Any highly-threaded CPU platform (Sun OpenSPARC UltraSPARC T series) help to provide hardware acceleration to the polling processes, reducing the amount of time proprietary CPU's normally spend on context switches, constantly pulling/pushing registers from/to slow memory.

While Network Management sees great performance strides with highly threaded hardware during 24x7x365 operations, not all areas are optimized. Two areas where highly threaded hardware (with slower individual thread performance) need improvement include: startup/shutdown time when the database needs to be loaded/dumped and post-discovery time when data needs to be consolidated with relationships built to all the other objects.

On some very large network topologies with extremely high (99.999) availability concerns - a slow startup time or slow post-discovery time is considered unacceptable - every minute counts in the case of software or hardware failure when protected with a High Availability kit.

Rock provides a mid-range position, between the $100K US$ super-threaded UltraSPARC T2+ SMP processors with high throughput (256 hardware threads) leveraging slower threads, and super-fast $1M US$ super-core'ed SPARC64 VII SMP (256 hardware threads) leveraging faster threads.

Rock's role in network management is clearly defined and beneficial.

Conclusion

Trade journalist writer Jon Stokes came to a diametrically opposite conclusion:

In the end, I can't say that I'm really sold on Sun's very aggressive use of speculative execution, but I will say that Rock is one of the most interesting and novel processors that I've seen in 10 years of covering this space. In its own way, it's every bit as exotic as IBM's Cell processor, but because all of that exoticism is hidden from the programmer it won't be nearly as difficult for developers to deal with.

I think this says it all.