Real World Max Simultaneous requests

Sun recommends this performance-tuning variable beset between 100 and 500.
To me that seems outdated. For a given hardware configuration what are people using for this value?
I�m looking at single processor with 2GB memory.

The maximum number of simultaneous requests is specified by the magnus.conf RqThrottle directive.
The ideal value for RqThrottle is determined more by the type of content served than by the hardware the web server runs on. If requests can be serviced quickly, you should use a lower number. If it takes a long time to service requests, you should use a higher number.
For example, if your web server just serves small GIF images, an RqThrottle value of 16 might yield the best performance. However, if your server executes complex database queries or streams movie files, an RqThrottle value of 1024 might be more appropriate.

Similar Messages

"Max simultaneous tempalate requests" is shown as "-1", set and restart yields unstable server

After CHF4 applied to CF901 (standard) and updated to run JRE7 on Windows server 2008, "Max simultaneous tempalate requests" is shown as "-1".
We checked the neo-runtime.xml and all of the limits look normal. We have no idea where does that "-1" come from.
We try setting it to a sensible value (i.e. 25) and then restart, and the CF server becomes very unstable. We couldn't even log into CF Administrator. The other live CF site would go down for a few minutes, and up again for the next few minutes. CPU activity is close to 0, RAM usage is stable. CF server logs shows nothing wrong, but http log did not get requests from IIS when it's down.
What's wrong?
Bug report filed: https://bugbase.adobe.com/index.cfm?event=bug&id=3528480

Not a good plan, IMHO.
Once that $cred object is created, the user can discover the password:
$cred.GetNetworkCredential().password
I think you'd be much better off creating a constrained, delegated remote session endpoint on a server somewhere, using the admin credentials as the RunAs account, and then creating some functions in the session to restart servers or services and limit the
session to just running those functions. You can add whatever logging you want as part of the functions, and audit using the Powershell event logs on the server.
http://blogs.technet.com/b/heyscriptingguy/archive/2014/04/02/build-constrained-powershell-endpoint-using-configuration-file.aspx
Give the users you want to be able to perform server and service restarts Execute permssions for the endpoint.
They won't have any knowledge of the admin account that's being used, and can't do anything you haven't given then a function for.
[string](0..33|%{[char][int](46+("686552495351636652556262185355647068516270555358646562655775 0645570").substring(($_*2),2))})-replace " "

Check Point 620 Wired Appliance: Real World Throughput and Experiences

I am considering a purchase of the Check Point 620 wired appliance. If you have used the 620, I would like to request some help:
1. What real-world throughput have you seen with all blades active? The internet speed at our site is 120 Mbps
2. What has your experience been -- good and bad? Anything notable?
Thanks!
This topic first appeared in the Spiceworks Community

I am considering a purchase of the Check Point 620 wired appliance. If you have used the 620, I would like to request some help:
1. What real-world throughput have you seen with all blades active? The internet speed at our site is 120 Mbps
2. What has your experience been -- good and bad? Anything notable?
Thanks!
This topic first appeared in the Spiceworks Community

Maximum Number of Simultaneous Requests

I want to update the Maximum Number of Simultaneous Requests
from 4 to 6 or 8. A restart of ColdFusion Application Server is
required for changes to take effect. Has anyone experienced
problems with the change not taking effect and/or causing the
service to not restart.
This has not happened. Being precautionary. The reason for
the question is that I made a change previously to the Max Heap
size from 1023 to 1535 in CF Admin, but when I attempted to restart
the CF service it would not. Finally had to change the Max Heap in
jvm.config back to 1023. Unsure why that would not accept it. The
variable is that the service is running with a non-admin user ID on
the server. Thoughts?

The prevailing wisdom from Adobe for simultaneous requests is
very wrong and inaccurate. First off, editing the simultaneous
requests in the CFAdmin is safe to do. Editing your JVM settings
with the CFAdmin is very dangerous on Linux because the CF Admin
code can mangle the xml file. I'm not sure if this is true on
Windows.
Now back to the simultaneous requests issue. If you have high
traffic and enough server processing power you can greatly increase
the request number. We currently run our CFMX 7.02 servers set to
100 simultaneous requests. And yes we've been maxed out at that
level. We see over 1.5 million page views per day on a single cf
server with only one instance of CF. As of today we switched to a
load balanced setup and split the load across two servers. The
reason we went load balanced is that we're expecting to more than
double our traffic. Anyways, the number of simultaneous requests
can be much higher than the 'General Wisdom' at Adobe.
Oh yeah, I almost forgot. I've seen the new setting for
simultaneous requests take effect with out having to restart CFMX.
Cheers,

Icreasing maximum number of Simultaneous requests

Hello
I am working on running a web based application , dployed on SunONE app server on a Windows 2000 machine. The goal is to test the application under stress condition. The test client sends 500 simultaneous requests, but I see the number of threads on the appservd process as 176 . I have already increased the max number of Simultaneous threads setting on the HttpServere. Is there any other thing that needs to be changed
thanks
-Aniruddha

the appserver uses a pool of threads for servicing client requests. it doesn't spawn a new thread for each request. after handling a request from a client, the same thread can handle a request from another client.
The following link (although it is written for the webserver, it also applies to the appserver) contains some more information that you may find useful. A similar document for the appserver is in the works.
http://docs.sun.com/source/816-5690-10/perf6.htm#18783
Thanks,
Arvind

Looking for real world effects site

Hi everybody
This is the first time ever using this forum, so please don't let me down.
I'm using FCS3 and doing some small jobs for a while now, and recently, I've been asked to film a music video, and the guy wants some professional effects, and i got scared since then.
here comes my request; I need a website or ... that show me then explain to me some real world effects that i can use it in the music video,
I'm really tired of looking and searching.
Thanks very much
Salah

Oho...so you want to go from dabbling in video with some small jobs, to doing high end effects like you see in THE MATRIX? Quite a leap. And you want to learn how to do this via online tutorials and the like?
Sigh.
like how to freeze frame while sweeping the camera about 180 degree around the objects (people).
Rent THE MATRIX and watch the EXTRA called BULLET TIME. See how they use 150 still cameras linked together to go off mere milliseconds apart. HIgh end visual effects guys with years of training figured that out.
i need someone to say this is a new tool in motion and you can use it to create this kind of effects ( real world, with people )
Not gonna happen because it doesn't exist. YOu want to have some application or plugin somehow be able to freeze someone, and spin them around, somehow seeing the OTHER SIDE of them, even though the one camera is on one side. Or, you can use 5-10 cameras all shooting that person, and use something like After Effects or other 3D program that stitches together that footage to make it do just that. Not simple...not by a longshot.
that's what I'm searching for. new professional stuff.
Professionals don't usually put their trade secrets online. Otherwise any old person could do what they did, and they wouldn't get paid the big bucks to figure it out, and do it on music videos and movies. That's why you don't find anything online.
Sorry to be snarky here...but I still wish you luck in figuring this out.
Shane

"The Real World"

Please bring more seasons of "The Real World" to the iTunes Music Store. I, as well as many other people, would shell out the 30 or so dollars for the older seasons!

user-to-user forum here. send requests to iTunes feedback.
JGG

ESATA 6g, USB 3 on PCI, real world

I am looking to buy a
CalDigit FASTA-6GU3
2 Port USB 3.0 & eSATA 6Gb/s Host Adapter
to put in my 2008 Mac Pro PCI slot to run an external Blu-Ray writer and hard drive enclosure off the eSATA ports.
Looking for info on what real-world Gps/Mbps the Mac Pro PCI slot can actually handle -- it is great if the card is 6Gbps eSATA & 5Gbps USB3.0 -- but the bottleneck is what the PCI bus can move (also limited by what the hadrdrive and writer can actually read/write).
I read on the reviews that the Mac Pro PCI bus can only perform tops 500Mbps (slightly above Firewire 400Mbps).
I also read the actual PCI card performance was limited to 250MBps due to its PLX chip design.
As USB2.0 is based on 480Mbps -- I have to wonder what all these numbers mean.
Can anyone put this into layman's perspective -- why put a 6Gps eSATA and USB 3.0 card in a PCI slot that is limited to 250Mbps?

Bytes will come to byte you if you think they are bits.
Kilobytes was the same issue. Or gigabits and bytes.
10 bits to a byte basically. sometimes 8 bits if there is no check bit.
Wikipedia is your friend, more so than Google. And there are dictionaries online of tech terms.
Some poeple come here talking about their MacPro except that they don't know what a 65lbs Mac is and my guess is there are a dozen notebook users asking.
The Sonnet E4P ~$250 w/ 4 x eSATA ports is nice for adding direct connect to SATA devices, more than enough bandwidth.
Your Mac has those four internal SATA2 drive bays. Those are all on a single shared controller bus, so while in theory you can get 275MB/sec max from each, there also is NOT full 1GB let alone 1.2 (1200MB/sec) but rather 750MB/sec - more than enough for most disk drives still. Only when you got to SSDs that it was more apparent.
Here is what your Mac's Serial ATA drives and bus are offering in bandwidth performance, sortof.
180MB/sec is about the max of 4TB enterprise drive, or WD 10K latest VelocityRaptors.
SATA2 SSD was about 275MB/sec also, and 450-500/550 (max, writes are often less with SSDs, reads tend to do very well, but it is the very high I/O's per second that they shine, along with lowest response time in seeks being 1/1000th - instead of being 4-12 milliseconds you are into less than 1.0 nanoseconds.
Geeks were happy with a Base2 numbering - binary 1's and 0's and base16 whereas Apple and others now wanted to change one gigabyte from being binary 1024MB and turn it into 1000MBs. The dumbing down of math, specs, and the i-ification of consumer devices. iDevice, iOS and consumers that never liked those 1024 binary and hex based numbering system.

Real world battery run time of new 15" mbp retina? (on-board graphics vs. video upgrade)

Considering a new Mac Book Pro 15" retina to use as a dedicated DAW in the studio. Primarily audio production work but would also do some video editing (Youtube videos, etc.) Will be primarily in studio but 10 - 15% in the field. (Hence the MBP rather than a desktop.) I'm looking at the top-end 15" retina maxed-out but not sure if I need or want the Nvidia GT750M graphics card.
My concern is battery run-time with the graphics card. I know Apple site claims 8 hours for the 15" retina. But their footnote states that's on a pre-production unit without the graphics upgrade. I've read multiple independent reviews of the version with graphics upgrade citing real world = 3 hour run times with "casual use". Yikes!
So I'm looking for real-world run time for the latest generation 15" retina with graphics upgrade vs. the Iris Pro option.   Anybody doing audio / video work with these two options care to share your experience? I don't think the DAW software (Logic Pro X) needs the graphics card and I'd be fine with the Iris Pro on-board graphics for audio work. Video editing would of course benefit from the card (2mg graphics memory.) It's only a $100 upgrade to add the card so it's not a financial question.   If real world is twice the run time without the graphics card I'll go for run time. But if it's, "4 hours Iris Pro and 3 hours Nvidia card" I'll probably go for the processing power and suffer.
Thanks!

Considering a new Mac Book Pro 15" retina to use as a dedicated DAW in the studio. Primarily audio production work but would also do some video editing (Youtube videos, etc.) Will be primarily in studio but 10 - 15% in the field. (Hence the MBP rather than a desktop.) I'm looking at the top-end 15" retina maxed-out but not sure if I need or want the Nvidia GT750M graphics card.
My concern is battery run-time with the graphics card. I know Apple site claims 8 hours for the 15" retina. But their footnote states that's on a pre-production unit without the graphics upgrade. I've read multiple independent reviews of the version with graphics upgrade citing real world = 3 hour run times with "casual use". Yikes!
So I'm looking for real-world run time for the latest generation 15" retina with graphics upgrade vs. the Iris Pro option.   Anybody doing audio / video work with these two options care to share your experience? I don't think the DAW software (Logic Pro X) needs the graphics card and I'd be fine with the Iris Pro on-board graphics for audio work. Video editing would of course benefit from the card (2mg graphics memory.) It's only a $100 upgrade to add the card so it's not a financial question.   If real world is twice the run time without the graphics card I'll go for run time. But if it's, "4 hours Iris Pro and 3 hours Nvidia card" I'll probably go for the processing power and suffer.
Thanks!

Making Effective Use of the Hybrid Cloud: Real-World Examples

May 2015
Explore
The Buzz from Microsoft Ignite 2015
NetApp was in full force at the recent Microsoft Ignite show in Chicago, and it was clear that NetApp's approach to hybrid cloud and Data Fabric resonated with the crowd. NetApp solutions such as NetApp Private Storage for Cloud are solving real customer problems.
Hot topics at the NetApp booth included:
OnCommand® Shift. A revolutionary technology that allows you to move virtual machines back and forth between VMware and Hyper-V environments in minutes.
Azure Site Recovery to NetApp Private Storage. Replicate on-premises SAN-based applications to NPS for disaster recovery in the Azure cloud.
Check out the following blogs for more perspectives:
Microsoft Ignite Sparks More Innovation from NetApp
ASR Now Supports NetApp Private Storage for Microsoft Azure
Four Ways Disaster Recovery is Simplified with Storage Management Standards
Introducing OnCommand Shift
SHIFT VMs between Hypervisors
Infront Consulting + NetApp = Success
Richard Treadway
Senior Director of Cloud Marketing, NetApp
Tom Shields
Senior Manager, Cloud Service Provider Solution Marketing, NetApp
Enterprises are increasingly turning to cloud to drive agility and closely align IT resources to business needs. New or short-term projects and unexpected spikes in demand can be satisfied quickly and elastically with cloud resources, spurring more creativity and productivity while reducing the waste associated with over- or under-provisioning.
Figure 1) Cloud lets you closely align resources to demand.
Source: NetApp, 2015
While the benefits are attractive for many workloads, customer input suggests that even more can be achieved by moving beyond cloud silos and better managing data across cloud and on-premises infrastructure, with the ability to move data between clouds as needs and prices change. Hybrid cloud models are emerging where data can flow fluidly to the right location at the right time to optimize business outcomes while providing enhanced control and stewardship.
These models fall into two general categories based on data location. In the first, data moves as needed between on-premises data centers and the cloud. In the second, data is located strategically near, but not in, the cloud.
Let's look at what some customers are doing with hybrid cloud in the real world, their goals, and the outcomes.
Data in the Cloud
At NetApp, we see a variety of hybrid cloud deployments sharing data between on-premises data centers and the cloud, providing greater control and flexibility. These deployments utilize both cloud service providers (CSPs) and hyperscale public clouds such as Amazon Web Services (AWS).
Use Case 1: Partners with Verizon for Software as a Service Colocation and integrated Disaster Recovery in the Cloud
For financial services company BlackLine, availability, security, and compliance with financial standards is paramount. But with the company growing at 50% per year, and periodic throughput and capacity bursts of up to 20 times baseline, the company knew it couldn't sustain its business model with on-premises IT alone.
Stringent requirements often lead to innovation. BlackLine deployed its private cloud infrastructure at a Verizon colocation facility. The Verizon location gives them a data center that is purpose-built for security and compliance. It enables the company to retain full control over sensitive data while delivering the network speed and reliability it needs. The colocation facility gives Blackline access to Verizon cloud services with maximum bandwidth and minimum latency. The company currently uses Verizon Cloud for disaster recovery and backup. Verizon cloud services are built on NetApp® technology, so they work seamlessly with BlackLine's existing NetApp storage.
To learn more about BlackLine's hybrid cloud deployment, read the executive summary and technical case study, or watch this customer video.
Use Case 2: Private, Nonprofit University Eliminates Tape with Cloud Integrated Storage
A private university was just beginning its cloud initiative and wanted to eliminate tape—and offsite tape storage. The university had been using Data Domain as a backup target in its environment, but capacity and expense had become a significant issue, and it didn't provide a backup-to-cloud option.
The director of Backup turned to a NetApp SteelStore cloud-integrated storage appliance to address the university's needs. A proof of concept showed that SteelStore™ was perfect. The on-site appliance has built-in disk capacity to store the most recent backups so that the majority of restores still happen locally. Data is also replicated to AWS, providing cheap and deep storage for long-term retention. SteelStore features deduplication, compression, and encryption, so it efficiently uses both storage capacity (both in the appliance and in the cloud) and network bandwidth. Encryption keys are managed on-premises, ensuring that data in the cloud is secure.
The university is already adding a second SteelStore appliance to support another location, and—recognizing which way the wind is blowing—the director of Backup has become the director of Backup and Cloud.
Use Case 3: Consumer Finance Company Chooses Cloud ONTAP to Move Data Back On-Premises
A leading provider of online payment services needed a way to move data generated by customer applications running in AWS to its on-premises data warehouse. NetApp Cloud ONTAP® running in AWS proved to be the least expensive way to accomplish this.
Cloud ONTAP provides the full suite of NetApp enterprise data management tools for use with Amazon Elastic Block Storage, including storage efficiency, replication, and integrated data protection. Cloud ONTAP makes it simple to efficiently replicate the data from AWS to NetApp FAS storage in the company's own data centers. The company can now use existing extract, transform and load (ETL) tools for its data warehouse and run analytics on data generated in AWS.
Regular replication not only facilitates analytics, it also ensures that a copy of important data is stored on-premises, protecting data from possible cloud outages. Read the success story to learn more.
Data Near the Cloud
For many organizations, deploying data near the hyperscale public cloud is a great choice because they can retain physical control of their data while taking advantage of elastic cloud compute resources on an as-needed basis. This hybrid cloud architecture can deliver better IOPS performance than native public cloud storage services, enterprise-class data management, and flexible access to multiple public cloud providers without moving data. Read the recent white paper from the Enterprise Strategy Group, “NetApp Multi-cloud Private Storage: Take Charge of Your Cloud Data,” to learn more about this approach.
Use Case 1: Municipality Opts for Hybrid Cloud with NetApp Private Storage for AWS
The IT budgets of many local governments are stretched tight, making it difficult to keep up with the growing expectations of citizens. One small municipality found itself in this exact situation, with aging infrastructure and a data center that not only was nearing capacity, but was also located in a flood plain.
Rather than continue to invest in its own data center infrastructure, the municipality chose a hybrid cloud using NetApp Private Storage (NPS) for AWS. Because NPS stores personal, identifiable information and data that's subject to strict privacy laws, the municipality needed to retain control of its data. NPS does just that, while opening the door to better citizen services, improving availability and data protection, and saving $250,000 in taxpayer dollars. Read the success story to find out more.
Use Case 2: IT Consulting Firm Expands Business Model with NetApp Private Storage for Azure
A Japanese IT consulting firm specializing in SAP recognized the hybrid cloud as a way to expand its service offerings and grow revenue. By choosing NetApp Private Storage for Microsoft Azure, the firm can now offer a cloud service with greater flexibility and control over data versus services that store data in the cloud.
The new service is being rolled out first to support the development work of the firm's internal systems integration engineering teams, and will later provide SAP development and testing, and disaster recovery services for mid-market customers in financial services, retail, and pharmaceutical industries.
Use Case 3: Financial Services Leader Partners with NetApp for Major Cloud Initiative
In the heavily regulated financial services industry, the journey to cloud must be orchestrated to address security, data privacy, and compliance. A leading Australian company recognized that cloud would enable new business opportunities and convert capital expenditures to monthly operating costs. However, with nine million customers, the company must know exactly where its data is stored. Using native cloud storage is not an option for certain data, and regulations require that the company maintain a tertiary copy of data and retain the ability to restore data under any circumstances. The company also needed to vacate one of its disaster-recovery data centers by the end of 2014.
To address these requirements, the company opted for NetApp Private Storage for Cloud. The firm placed NetApp storage systems in two separate locations: an Equinix cloud access facility and a Global Switch colocation facility both located in Sydney. This satisfies the requirement for three copies of critical data and allows them to take advantage of AWS EC2 compute instances as needed, with the option to use Microsoft Azure or IBM SoftLayer as an alternative to AWS without migrating data. For performance, the company extended its corporate network to the two facilities.
The firm vacated the data center on schedule, a multimillion-dollar cost avoidance. Cloud services are being rolled out in three phases. In the first phase, NPS will provide disaster recovery for the company's 12,000 virtual desktops. In phase two, NPS will provide disaster recover for enterprise-wide applications. In the final phase, the company will move all enterprise applications to NPS and AWS. NPS gives the company a proven methodology for moving production workloads to the cloud, enabling it to offer new services faster. Because the on-premises storage is the same as the cloud storage, making application architecture changes will also be faster and easier than it would be with other options. Read the success story to learn more.
NetApp on NetApp: nCloud
When NetApp IT needed to provide cloud services to its internal customers, the team naturally turned to NetApp hybrid cloud solutions, with a Data Fabric joining the pieces. The result is nCloud, a self-service portal that gives NetApp employees fast access to hybrid cloud resources. nCloud is architected using NetApp Private Storage for AWS, FlexPod®, clustered Data ONTAP and other NetApp technologies. NetApp IT has documented details of its efforts to help other companies on the path to hybrid cloud. Check out the following links to lean more:
Hybrid Cloud: Changing How We Deliver IT Services [blog and video]
NetApp IT Approach to NetApp Private Storage and Amazon Web Services in Enterprise IT Environment [white paper]
NetApp Reaches New Heights with Cloud [infographic]
Cloud Decision Framework [slideshare]
Hybrid Cloud Decision Framework [infographic]
See other NetApp on NetApp resources.
Data Fabric: NetApp Services for Hybrid Cloud
As the examples in this article demonstrate, NetApp is developing solutions to help organizations of all sizes move beyond cloud silos and unlock the power of hybrid cloud. A Data Fabric enabled by NetApp helps you more easily move and manage data in and near the cloud; it's the common thread that makes the uses cases in this article possible. Read Realize the Full Potential of Cloud with the Data Fabric to learn more about the Data Fabric and the NetApp technologies that make it possible.
Richard Treadway is responsible for NetApp Hybrid Cloud solutions including SteelStore, Cloud ONTAP, NetApp Private Storage, StorageGRID Webscale, and OnCommand Insight. He has held executive roles in marketing and engineering at KnowNow, AvantGo, and BEA Systems, where he led efforts in developing the BEA WebLogic Portal.
Tom Shields leads the Cloud Service Provider Solution Marketing group at NetApp, working with alliance partners and open source communities to design integrated solution stacks for CSPs. Tom designed and launched the marketing elements of the storage industry's first Cloud Service Provider Partner Program—growing it to 275 partners with a portfolio of more than 400 NetApp-based services.
Quick Links
Tech OnTap Community
Archive
PDF

Dave:
"David Scarani" <[email protected]> wrote in message
news:3ecfc046$[email protected]..
>
I was looking for some real world "Best Practices" of deploying J2EEapplications
into a Production Weblogic Environment.
We are new at deploying applications to J2EE application servers and arecurrently
debating 2 methods.
1) Store all configuration (application as well as Domain configuration)in properties
files and Use Ant to rebuild the domain everytime the application isdeployed.
I am just a WLS engineer, not a customer, so my opinions have in some
regards little relative weight. However I think you'll get more mileage out
of the fact that once you have created your config.xml, checking it into src
control, versioning it. I would imagine that application changes are more
frequent than server/domain configuration so it seems a little heavy weight
to regenerate the entire configuration everytime an application is
deployed/redeployed. Either way you should check out the wlconfig ant task.
Cheers
mbg
2) Have a production domain built one time, configured as required andalways
up and available, then use Ant to deploy only the J2EE application intothe existing,
running production domain.
I would be interested in hearing how people are doing this in theirproduction
environments and any pros and cons of one way over the other.
Thanks.
Dave Scarani

Real World Adobe Photoshop CS3 (Real World)

Real World Adobe Illustrator CS3 (Real World) - Mordy
Golding;
Real World Adobe Photoshop CS3 (Real World) - David Blatner;
these books are in the UPPER LEVEL than "classroom in a book"
series ?

> but the part about DNG has convinced me to dive deeper in it and give it a go
When working in a Bridge/Camera Raw/Photoshop workflow, I tend to ingest the actual native raw files, do initial selects and gross edits and basic metadata work via templates and THEN do the conversion to DNG. I'll use the DNG as my working files and the original raws as an archive. I tend to do this more with studio shoots. I tend to use Lightroom when I'm on the road.
When working in Lightroom first, I tend to ingest and convert to DNG upon ingestion (when in the road working on a laptop) while using the backup copyusually working on a pair of external FW drives one for working DNG files and 1 for BU of the original raws. Then, when I get back to the studio I make sure I write to XMP and export the new shoot as a catalog and import into my studio copy of Lightroom. Then I'll also cache the newly imported images in Bridge as well so I can get at the image either in Bridge or Lightroom.
It's a bit of a chore now since I do work in Camera Raw a lot (well, DOH, I had to to do the book!) but I also keep all my digital files in a Lightroom catalog which is now up to about 74K...
Then, depending on what I'll need to do, I'll either work out of LR or Bridge/Camera Raw...
If I'm doing a high-end final print, I generally process out of Camera Raw as a Smart Object and stack multiple layers of CR processed images...if I'm working on a batch of images I'll work out of Lightroom since the workflow seems to suit me better.
In either event, I've found DNG to be better than native raws with sidecar files.

RAID test on 8-core with real world tasks gives 9% gain?

Here are my results from testing the software RAID set up on my new (July 2009) Mac Pro. As you will see, although my 8-core (Octo) tested twice as fast as my new (March 2009) MacBook 2.4 GHz, the software RAID set up only gave me a 9% increase at best.
Specs:
Mac Pro 2x 2.26 GHz Quad-core Intel Xeon, 8 GB 1066 MHz DDR3, 4x 1TB 7200 Apple Drives.
MacBook 2.4 GHz Intel Core 2 Duo, 4 GB 1067 MHz DDR3
Both running OS X 10.5.7
Canon Vixia HG20 HD video camera shooting in 1440 x 1080 resolution at “XP+” AVCHD format, 16:9 (wonderful camera)
The tests. (These are close to my real world “work flow” jobs that I would have to wait on when using my G5.)
Test A: import 5:00 of video into iMovie at 960x540 with thumbnails
Test B: render and export with Sepia applied to MPEG-4 at 960x540 (a 140 MB file) in iMovie
Test C: in QuickTime resize this MPEG-4 file to iPod size .m4v at 640x360 resolution
Results:
Control: MacBook as shipped
Test A: 4:16 (four minutes, sixteen seconds)
Test B: 13:28
Test C: 4:21
Control: Mac Pro as shipped (no RAID)
Test A: 1:50
Test B: 7:14
Test C: 2:22
Mac Pro config 1
RAID 0 (no RAID on the boot drive, three 1TB drives striped)
Test A: 1:44
Test B: 7:02
Test C: 2:23
Mac Pro config 2
RAID 10 (drives 1 and 2 mirrored, drives 3 and 4 mirrored, then both mirrors striped)
Test A: 1:40
Test B: 7:09
Test C: 2:23
My question: Why am I not seeing an increase in speed on these tasks? Any ideas?
David
Notes:
I took this to the Apple store and they were expecting 30 to 50 per cent increase with the software RAID. They don’t know why I didn’t see it on my tests.
I am using iMovie and QuickTime because I just got the Adobe CS4 and ran out of cash. And it is fine for my live music videos. Soon I will get Final Cut Studio.
I set up the RAID with Disk Utility without trouble. (It crashed once but reopened and set up just fine.) If I check back it shows the RAID set up working.
Activity Monitor reported “disk activity” peaks at about 8 MB/sec on both QuickTime and iMovie tasks. The CPU number (percent?) on QT was 470 (5 cores involved?) and iMovie was 294 (3 cores involved?).
Console reported the same error for iMovie and QT:
7/27/09 11:05:35 AM iMovie[1715] Error loading /Library/Audio/Plug-Ins/HAL/DVCPROHDAudio.plugin/Contents/MacOS/DVCPROHDAudio: dlopen(/Library/Audio/Plug-Ins/HAL/DVCPROHDAudio.plugin/Contents/MacOS/DVCPROHD Audio, 262): Symbol not found: _keymgr_get_per_threaddata
Referenced from: /Library/Audio/Plug-Ins/HAL/DVCPROHDAudio.plugin/Contents/MacOS/DVCPROHDAudio
Expected in: /usr/lib/libSystem.B.dylib

The memory controllers, one for each cpu, means that you need at least 2 x 2GB on each bank. If that is how Apple set it up, that is minimal and the only thing I would do now with RAM is add another 2 x 2GB. That's all. And get you into triple channel bandwidth.
It could be the make and model of your hard drives. If they are seagate then more info would help. And not all drives are equal when it comes to RAID.
Are you new to RAID or something you've been doing? seems you had enough to build 0+1 and do some testing. Though not pleased, even if it works now, that it didn't take the one time.
Drives - and RAIDs - improve over the first week or two - which, before commiting good data to them - is the best time to torture, run them ragged, use Speedtools to break them in, loosen up the heads, scan for media errors, and run ZoneBench (and with 1TB, partition each drive into 1/4ths).
If Drive A is not identical to B, then they may deal with an array even worse. And no two drives are purly identical, some vary more than others, and some are best used in hardware RAID controller environments.
Memory: buying in groups of three. okay. But then adding 4 x 4GB? So bank A with 4 x 2GB and B with twice as much memory. On Mac Pro, 4 DIMMs on a bank you get 70% bandwidth, it drops down from tri-channel to dual-channel mode.
I studied how to build or put together a PC for over six months, but then learned more in the month (or two) after I bought all the parts, found what didn't work, learned my own short-comings, and ended up building TWO - one for testing, other for backup system. And three motherboards (the best 'rated' also had more trouble with BIOS and fans, the cheap one was great, the Intel board that reviewers didn't seem to "gork" actually has been the best and easiest to use and update BIOS). Hands on wins 3:1 versus trying to learn by reading for me, hands-on is what I need to learn. Or take car or sailboat out for drive, spin, see how it fares in rough weather.
I buy an Apple system bare bones, stock, or less, then do all the upgrades on my own, when I can afford to, gradually over months, year.
Each cpu needs to be fed. So they each need at least 3 x 1GB RAM. And they need raw data fed to RAM and cpu from disk drives. And your mix of programs will each behave differently. Which is why you see Barefeats test with Pro Apps, CINEBENCH, and other apps or tools.
What did you read or do in the past that led you to think you need RAID setup, and for how it would affect performance?
Photoshop Guides to Performance:
http://homepage.mac.com/boots911/.Public/PhotoshopAccelerationBasics2.4W.pdf
http://kb2.adobe.com/cps/401/kb401089.html
http://www.macgurus.com/guides/storageaccelguide.php
4-core vs 8-core
http://www.barefeats.com/nehal08.html
http://www.barefeats.com/nehal03.html

EJB 3.0 in a real world open source project. Great for coding reference!

If you are interested in seeing EJB 3.0 implemented in a real world project (not just examples) or if you are interested in learning how to use them I suggest you to take a look a the open source project Overactive Logistics.
It has been written totally using EJB 3.0 (session and entity beans) I found it very helpful in solving several technical situations I was facing.
You can get more information at:
http://overactive.sourceforge.net

Thanks for the ponter, I will check it out.
hth,
Sean

Character Styles in the Real World

Rick:
Thanks for your efforts, and let me add my Amen to both
subjects (on file locations and on Character styles).
My real-world use of Character styles is a combination usage
of Paragraph and Character styles for Notes: I have a Paragraph
style called Note, which simply adds margins of .15in Left, 10pt
Top, and 8pt Bottom. Within this paragraph style, multiple labels
announce the type of Note with the use of Character styles
NoteLabel (Navy), RecommendLabel (Teal), CAUTIONLabel (Purple), and
WARNINGLabel (Red).
This way, you can change the color of one or more labels
without worrying about the paragraph settings (or vice versa).
Also, when placing a Note inside a table cell (which might
have limited horizontal space, especially with three or four
columns), we still use the "Label" character styles but
without the Notes paragraph style. This still sets off the
text visually, without adding unnecessary extra vertical space.
Thanks again, Rick!
Leon

I can tell you about two sites.
1. A system which allocates and dispatches crews, trucks, backpack hoses, spare socks, etc to bushfires (wildfires to you). It operates between two Government departments here in Australia. Each of those despatchable items is a remote object and there have been up to 50,000 active in the system at a time during the hot summer months. This is a large and life-critical system.
2. A monitoring system for cable TV channels. A piece of hardware produces a data stream representing things like channel utilization, error rates, delay, etc and this is multiplexed via RMI to a large number of operator consoles. Again this is a major and business-critical system.
And of course every J2EE system in existence uses RMI internally, albeit almost entirely RMI/IIOP.

PI Implementation Examples - Real World Scenarios

Hello friends,
In the near future, I'm going to give a presentation to our customers on SAP PI. To convince them to using this product, I need examples from the real world that are already implemented successfully.
I have made a basic search but still don't have enough material on the topic, I don't know where to look actually. Could you post any examples you have at hand? Thanks a lot.
Regards,
Gökhan

Hi,
Please find here with you the links
SAP NetWeaver Exchange Infrastructure Business to Business and Industry Standards Support (2004)
https://www.sdn.sap.com/irj/sdn/go/portal/prtroot/docs/library/uuid/90052f25-bc11-2a10-ad97-8f73c999068e
SAP Exchange Infrastructure 3.0: Simple Use Cases
https://www.sdn.sap.com/irj/sdn/go/portal/prtroot/docs/library/uuid/20800429-d311-2a10-0da2-d1ee9f5ffd4f
Exchange Infrastructure - Integrating Heterogeneous Systems with Ease
https://www.sdn.sap.com/irj/sdn/go/portal/prtroot/docs/library/uuid/1ebea490-0201-0010-faad-a32dd753d009
SAP Network Blog: Re-Usable frame work in XI
/people/sravya.talanki2/blog/2006/01/10/re-usable-frame-work-in-xi
SAP NetWeaver in the Real World, Part 1 - Overview
https://www.sdn.sap.com/irj/sdn/go/portal/prtroot/docs/library/uuid/20456b29-bb11-2a10-b481-d283a0fce2d7
SAP NetWeaver in the Real World, Part 3 - SAP Exchange Infrastructure
https://www.sdn.sap.com/irj/sdn/go/portal/prtroot/docs/library/uuid/3172d290-0201-0010-2b80-c59c8292dcc9
SAP NetWeaver in the Real World, Part 3 - SAP Exchange Infrastructure
https://www.sdn.sap.com/irj/sdn/go/portal/prtroot/docs/library/uuid/9ae9d490-0201-0010-108b-d20a71998852
SAP NetWeaver in the Real World, Part 4 - SAP Business Intelligence
https://www.sdn.sap.com/irj/sdn/go/portal/prtroot/docs/library/uuid/1f42d490-0201-0010-6d98-b18a00b57551
Real World Composites: Working With Enterprise Services
https://www.sdn.sap.com/irj/sdn/go/portal/prtroot/docs/library/uuid/d0988960-b1c1-2a10-18bd-dafad1412a10
Thanks
Swarup

Real World Max Simultaneous requests

Similar Messages

Maybe you are looking for