Radified Community Forums - Intelligent File Access Acceleration Sequencing

Brian

Demigod

Offline

Posts: 6345
NSW, Australia

Back to top

	Re: Intelligent File Access Acceleration Sequencin Reply #60 - Dec 5^th, 2006 at 2:32pm Ivanov, Quote: Hypothetically what would be pros & cons of having & not having a page file, if the total commit charge is around 700MB & the installed physical memory is 2 GB? http://aumha.org/win5/a/xpvm.htm
	IP Logged

Pleonasm

Übermensch

Offline

Posts: 1619

Back to top

	Re: Intelligent File Access Acceleration Sequencin Reply #61 - Dec 6^th, 2006 at 9:36am Readers of this thread may be interested in the free disk defragmentation tool DirMS.
	ple • o • nasm n. “The use of more words than are required to express an idea” IP Logged

Pleonasm

Übermensch

Offline

Posts: 1619

Back to top

Re: Intelligent File Access Acceleration Sequencin
Reply #62 - Dec 6^th, 2006 at 9:45am

Diskeeper 2007 Pro Premier reports an actual I-FAAST performance gain of 18% on my PC, together with these two statistics:
Most Accessed Throughput: 52,831 KB/s
Least Accessed Throughput: 34,856 KB/s

Based on the throughput statistics, the gain delivered by I-FAAST is 52% and not 18%.

Observations and comments are welcome.

ple • o • nasm n. “The use of more words than are required to express an idea”

IP Logged

NightOwl

Radministrator

Offline

"I tought I saw a puddy
tat..."

Posts: 5826
Olympia, WA--Puget Sound--USA

Back to top

Re: Intelligent File Access Acceleration Sequencin
Reply #63 - Dec 6^th, 2006 at 10:28am

Pleonasm

What were those figures *before* I-FAAST?

I'm guessing--*Most Accessed* are probably in the HDD *cache* (? on board memory chip (? RAM) inside the HDD) more often than not, but *Least Accessed* need to be brought in from the disk platters because they have not been accessed recently.

Possibly the difference between *burst* through put vs *sustained* through put?

____________________________________________________________________________________________

No question is stupid ... but, possibly the answers are Wink

!

IP Logged

Pleonasm

Übermensch

Offline

Posts: 1619

Back to top

Re: Intelligent File Access Acceleration Sequencin
Reply #64 - Dec 6^th, 2006 at 11:29am

Greetings, NightOwl.

My interpretation of the throughput rates in Reply #62 are that they reflect the reorganization of files that occurred as a consequence of the operation of I-FAAST: i.e., that I-FAAST placed the most accessed files on that portion of the disk surface where the throughput is highest. Those throughput rates are not reported, as I recall, by Diskeeper 2007 Professional (which lacks the I-FAAST capability found only in the 2007 Pro Premier version).

Because these statistics are displayed by Diskeeper under the heading “I-FAAST Performance Gains and Throughput Rates” and all on one single horizontal line, I do not believe that are intended to show general performance characteristics of the hardware (e.g., burst versus sustained, or RAM cache versus magnetic media retrieval).

Maybe the I-FAAST performance gain of 18% is computed by comparing the throughput of the most accessed files (52,831 KB/s) to the mean throughput of the drive overall? Maybe it is computed by weighting the increased file access speed by the relative frequency with which those specific files are read?

From a marketing perspective, it seems odd (but good) that Diskeeper would choose a computation method for the I-FAAST performance gain statistic that is perhaps unnecessarily “conservative.” I have sent a request to Diskeeper Technical Support to learn more about how they compute the percent performance gain, and will post what I learn.

Brian, what are the I-FAAST statistics you are seeing for the "Actual Performance Gain" (%) and the Most/Least Accessed Throughput rates (KB/s)?

ple • o • nasm n. “The use of more words than are required to express an idea”

IP Logged

Ivanov

Gnarly

Offline

I love YaBB 1G - SP1!

Posts: 43

Back to top

	Re: Intelligent File Access Acceleration Sequencin Reply #65 - Dec 6^th, 2006 at 12:42pm Pleonasm, Sorry about the misunderstanding. I know boot time is manual by default & has to be started. But when IFAAST is on, Boot time will not work manually till IFAAST is unchecked.......?? Regards!
	Ivanov. IP Logged

Pleonasm

Übermensch

Offline

Posts: 1619

Back to top

Re: Intelligent File Access Acceleration Sequencin
Reply #66 - Dec 6^th, 2006 at 5:13pm

Ivanov, I am not aware that having I-FAAST enabled prohibits the use of the Boot-Time Defragmentation capability. If you find that occurring, then – of course – it’s a simple matter to disable I-FAAST, run the Boot-Time Defragmentation, and then enable I-FAAST again.

Are you using Diskeeper 2007 Pro Premier or Diskeeper 10?

ple • o • nasm n. “The use of more words than are required to express an idea”

IP Logged

Ivanov

Gnarly

Offline

I love YaBB 1G - SP1!

Posts: 43

Back to top

	Re: Intelligent File Access Acceleration Sequencin Reply #67 - Dec 7^th, 2006 at 12:30am Pleonasm, I am using DK Pro Premier Version: 10.0.608.0. That’s what I do if I want to run Boot time, I disable/uncheck I-FAAST, and after Boot time has done its job enable I-FAAST again. Regards!
	Ivanov. IP Logged

Brian

Demigod

Offline

Posts: 6345
NSW, Australia

Back to top

Re: Intelligent File Access Acceleration Sequencin
Reply #68 - Dec 9^th, 2006 at 2:01am

Pleonasm,

There is nothing exciting to report. Subjectively we feel the computer behaves the same with ver 9 or ver 11 (Pro/Premier) installed. Times to open and run various programs and access data were the same for both versions.

I couldn’t assess I-FAAST probably because my partitions were too small. The C: drive was 6 GB and I-FAAST reported 0% Estimated Performance Gain and an Actual Performance Gain of -----

The Program partition (5 GB) reported 2% Estimated Performance Gain and an Actual Performance Gain of -----

The Data partition (50 GB) reported 20% Estimated Performance Gain but also reported I-FAAST has not run.

I chose Automatic Defrag for all partitions and for I-FAAST I did choose Allow Diskeeper to enable or disable.

The Most/Least Accessed Throughput rates (KB/s) for the C: drive were 58843 and 58642.

I’ve saved money by running the trial app.

IP Logged

Pleonasm

Übermensch

Offline

Posts: 1619

Back to top

Re: Intelligent File Access Acceleration Sequencin
Reply #69 - Dec 9^th, 2006 at 11:18am

Brian, sorry to hear that I-FAAST is not able to deliver a performance gain on your PC. Because I-FAAST isn't running on your system, I would expect (and your data confirm) that there is little difference between Diskeeper 9 and Diskeeper 2007.

The Most/Least Accessed Throughput rates (KB/s) on your PC were nearly identical, in contrast to the huge 52% difference that I observed on my PC (see Reply #62).

Based on a combination of your experience and mine, I would offer the conclusion that (A) I-FAAST isn't for every PC, but (B) when it works, it works very well.

ple • o • nasm n. “The use of more words than are required to express an idea”

IP Logged

Pleonasm

Übermensch

Offline

Posts: 1619

Back to top

	Re: Intelligent File Access Acceleration Sequencin Reply #70 - Dec 9^th, 2006 at 2:50pm Brian, you reported a problem with Diskeeper 9 when used in conjunction with the creation of incremental recovery points by Ghost 10 (see Reply #7). Did you find that the issue also existed when using Diskeeper 2007 or not?
	ple • o • nasm n. “The use of more words than are required to express an idea” IP Logged

Brian

Demigod

Offline

Posts: 6345
NSW, Australia

Back to top

	Re: Intelligent File Access Acceleration Sequencin Reply #71 - Dec 10^th, 2006 at 2:04am Pleo, I don't run incremental images on my son's computer.
	IP Logged

Ivanov

Gnarly

Offline

I love YaBB 1G - SP1!

Posts: 43

Back to top

Re: Intelligent File Access Acceleration Sequencin
Reply #72 - Dec 10^th, 2006 at 7:26am

Pleonasm,

I would like to add to your conclusions about I-FASST, based on my personal & practical experience.

I agree that I-FAAST isn’t for every pc.

It’s just like the same disease manifests differently in patients & at times the book picture is rarely there. Apparently all humans may look alike but the genetics / internals differ…..

One drug would work instantly on controlling symptoms in a particular case whilst in a similar case that drug wouldn’t be that effective….

Coming back to I-FAAST its effectiveness mainly depends on 2 things:

Firstly, the size of the drive/partition.

Secondly what sort of data that drive contains.

Partition/Drive Size is the most important in my view.

As for most users who partition their primary drive into 3/4 partitions, it’s of no good.

Personally, I feel PD is better in such cases & DK is great when stand alone has large sized primary drives/ partitions.

DK & I-FAAST work great for such stand alone & servers…….

I could be totally wrong but these are my personal thoughts……

Regards!

Ivanov.

IP Logged

Ivanov

Gnarly

Offline

I love YaBB 1G - SP1!

Posts: 43

Back to top

	Re: Intelligent File Access Acceleration Sequencin Reply #73 - Dec 10^th, 2006 at 2:10pm For your views:
	Ivanov. IP Logged

Pleonasm

Übermensch

Offline

Posts: 1619

Back to top

Re: Intelligent File Access Acceleration Sequencin
Reply #74 - Apr 26^th, 2007 at 10:34am

Disk defragmentation improves the performance of backup operations, of course; but note the potential magnitude of the impact:

Quote:

Testing has found that the time required to back up data from a typical hard drive volume to backup devices can be decreased and the backup data transfer rate increased by a defrag job prior to backup. Furthermore, directory consolidation has a direct additional benefit in backup and data transfer rates. The study showed improvements in backup performance up to 69% (decrease in total back up time)—less than one half of the time for the fragmented system. Similarly, backup data transfer rates increased up to 69% over that for the fragmented state—nearly 1.7 times as fast.

Combining the benefits of directory consolidation with I-FAAST (Intelligent File Access Acceleration Sequencing Technology) improved backup data transfer rates and backup times in the network disk backup trials for ARCServe by a total of 70%.

Source: Defrag Study Indicates Increased Backup Performance

ple • o • nasm n. “The use of more words than are required to express an idea”

IP Logged

	Welcome, Guest. Please Login