|
If boot/app startup was the only benefit, you're right an SSD wouldn't add much, but since blocking IO is common in many apps almost everything becomes more responsive. Measuring system responsiveness isn't anywhere as easy as benching boot/app startup; but it's the responsiveness gains that gave me the real wow.
If you have to choose between enough dram to keep all your working data in memory or an SSD obviously the former is more important; but windows vista and later don't report IO operations as complete until they've been flushed from the cache to the disk and dropping the latency from each write from ~10ms to a few microseconds actually is noticeable in normal useage. EDIT: I didn't mention this in my original post because "more ram than I'll actually use" has IMO been implicit in any custom configured PC for the last few years. Only crappy retail configs skimp there any longer.
Unless you're CPU bound however I'd put an SSD as more useful than a faster CPU because it eliminates the most generally apparent performance bottleneck; when random IO trashes a magnetic HD.
About 4 years ago I blew about $6-700 on a hardware raid controller and 5 HDs to try and boost performance; outside of benchmarks I never noticed a difference. The responsiveness gain going from an i7@3.85ghz with a single HD to the same system with an SSD is almost as dramatic as the difference between the same system with an HD and my netbook with an HD.3x12=36
2x12=24
1x12=12
0x12=18
|
|
|
|
|
Dan Neely wrote: vista and later don't report IO operations as complete until they've been flushed from the cache to the disk
that is new to me. Are tou saying they've changed that since XP? or was XP the same? do you have a link to such info? I'm eager to read more about it.
Dan Neely wrote: I'd put an SSD as more useful than a faster CPU
I agree. I'd never go for the fastest CPU, the CPU typically isn't the bottleneck (except when trying to compute on a netbook of course).
|
|
|
|
|
Luc Pattyn wrote: that is new to me. Are tou saying they've changed that since XP? or was XP the same? do you have a link to such info? I'm eager to read more about it.
It was a change for vista. Look for MS articles about why reported file copy performance in vista is slower than what XP reported. XP reporting that it was done before the flushing was part of it. The other part was that they optimized the algorithm used for copies on crappy wifi networks with high latencies and packet loss figuring that in the real world large gains in the bad case were worth a small hit in the best case. The world of internet ranting disagreed vocally with that assessment and changes were again made in SP1 to boost the bestcase performance than internet ranters used for their benchmark graphs.  3x12=36
2x12=24
1x12=12
0x12=18
|
|
|
|
|
yeah, I start to recall that now. Thanks.
|
|
|
|
|
| Has anyone used ram disk software?? What I have been thinking is maybe loading it with ram and using ram disk software to keep everything I am working on in memory. Humble Programmer
|
|
|
|
|
I realize that I will have a big improvement with OS... everything I have read supports that but VS is what I want to really perform fast because that is what I use all day long and I have read things that say that there is little improvement and sometimes a decrease in performance because of all of the small files that have to be used in VS. I guess I am wondering does anyone know if solid state handles small files any better than before because much of the info I have read is dated so I want to know has anyone see it greatly improve the performance of VS. Also has anyone ever heard of a spindle hard drive that is faster than 7200 in a laptop.
Update I am looking at a Dell m6500 with 16GB of ram i7 quad 2.0ghz.
I currently have a Dell Vostro 1700 4GB ram intel core 2 duo 2.4ghz I know my bottle neck right now is my hard drive it is only 5400rpm which I did not know when it was ordered.
This computer is not bad I just want faster compile times and I also am fixing to start doing all dev in VMWare.Humble Programmer
|
|
|
|
|
Random IO is where SSDs have the biggest lead over magnetic drives. In sequential IO the gain for an SSD is only about 2x that of a magnetic HD. The fastest SSD controllers on the market are upto ~150x faster in random writes.
http://www.anandtech.com/storage/showdoc.aspx?i=3747&p=10[^]3x12=36
2x12=24
1x12=12
0x12=18
|
|
|
|
|
I've used dual socket workstation with 4-8 cores since 2005. The main benefit is greater stability and less noticeable interference from anti-virus. If your code has very few page faults and anti-virus isn't sucking up your cpu, you may actually have better performance with the fastest dual core Xeon or Core i7 cpu's unless you do a serious amount of computation. With VMWare, you definitely want to have a minimum of 8GB of RAM. If you can install two hard drives, then you can use two identical hard drives with RAID 0 striping, or a small SSD for programs and 500+ GB 7200 RPM SATA hard drive for your data.
modified on Monday, April 5, 2010 2:09 AM
|
|
|
|
|
Can only talk for myself now, but getting an SSD is one of the best things I ever done.
If it was my choice I'd get an SSD, a dual core processor and as much RAM as your budget allows possible."When did ignorance become a point of view" - Dilbert
|
|
|
|
|
Whether it's busy doing something or not, it just powers off (read: no graceful shutdown) right then and there. Can a bad HDD cause this, or would it be something on the mobo (e.g., fan)? The machine is at least four years old, but probably older.
- DC"One man's wage rise is another man's price increase." - Harold Wilson
"Fireproof doesn't mean the fire will never come. It means when the fire comes that you will be able to withstand it." - Michael Simmons
"Man who follows car will be exhausted." - Confucius
modified on Monday, March 8, 2010 11:13 PM
|
|
|
|
|
It can be bad memory (RAM, cache, HDD, Graphic card).
A bad solder, can happen basically anywhere. This is usually but not always temperature dependent. The power supply is extra sensitive due to higher temperature differences.
A bad fan, is at least easy to check.
Oxidized contacts on cables. Also usually temperature dependent.
But from the description I'd put my money on the PSU, you do say that the machine POWERS OFF, not that it crashes."When did ignorance become a point of view" - Dilbert
|
|
|
|
|
Jörgen Andersson wrote: ...you do say that the machine POWERS OFF, not that it crashes.
Correct."One man's wage rise is another man's price increase." - Harold Wilson
"Fireproof doesn't mean the fire will never come. It means when the fire comes that you will be able to withstand it." - Michael Simmons
"Man who follows car will be exhausted." - Confucius
|
|
|
|
|
DavidCrow wrote: Whether it's busy doing something or not, it just powers off (read: no graceful shutdown) right then and there. Can a bad HDD cause this, or would it be something on the mobo (e.g., fan)?
I'd go with over heating. I go into the BIOS and look at the CPU and Case temperatures and see if they are on the high side. If they are from just runningn in the BIOS, when Windows is up it will be hotter. If your lucky the BIOS will have a setting to shutdown the machine when it gets to a certain temperature, if not it will just shutdown when the CPU can't cope.
If it is heat, poull off the heatsink, use mineral turpentine to reemove the old thermal paste from the heatsink and CPU, then Methylated Spirits to remove the Mineral Turpentine (using a cloth of course, don't pour it in. Let it dry, add new thermal paste, replace heatsink, fire it up and see if it hasn't improved.
I would also clean out the dirt and dust from the heatsink, case and fans as this can clog up and heat up a system. I use an air compressor though many advise not too.Michael Martin
Australia
"I controlled my laughter and simple said "No,I am very busy,so I can't write any code for you". The moment they heard this all the smiling face turned into a sad looking face and one of them farted. So I had to leave the place as soon as possible."
- Mr.Prakash One Fine Saturday. 24/04/2004
|
|
|
|
|
My computer BSOD'd with UNMOUNTABLE_BOOT_VOLUME. Nothing that I've read on the Internet has worked in getting access to the HDD. If I try and boot from a CD, the machine acts as if no CD is present in the drive. If I hook the HDD up to a SATA/USB adapter, the host computer freezes. I'm at a loss as to what to try.
I've never tried anything like it before, but can the HDD be mounted on a Linux/Ubuntu system somehow?
Thanks.
- DC
[edit]
After installing the latest version of Ubuntu (9.x) and enabling NTFS writing on external devices, it's working. It should finish copying to the backup drive in roughly 62 hours from now (just about the time I get to work on Monday)!
[/edit]"One man's wage rise is another man's price increase." - Harold Wilson
"Fireproof doesn't mean the fire will never come. It means when the fire comes that you will be able to withstand it." - Michael Simmons
"Man who follows car will be exhausted." - Confucius
modified on Monday, March 8, 2010 2:42 PM
|
|
|
|
|
DavidCrow wrote: If I try and boot from a CD, the machine acts as if no CD is present in the drive
Maybe a stupid question, but have you set the bootorder in the BIOS to boot from the CD first?
DavidCrow wrote: I've never tried anything like it before, but can the HDD be mounted on a Linux/Ubuntu system somehow?
It's certainly worth trying, but my guess is that your HDD is fried"When did ignorance become a point of view" - Dilbert
|
|
|
|
|
Jörgen Andersson wrote: Maybe a stupid question, but have you set the bootorder in the BIOS to boot from the CD first?
Yes.
Jörgen Andersson wrote: It's certainly worth trying, but my guess is that your HDD is fried
I've got it mounted on a Ubuntu system, and it's reading fine. I now have to figure out how to back up the data (roughly 97GB) to an external USB HDD. Are these external HDDs preformatted? If so (e.g., FAT32), does that mean files written by Ubuntu can be read by Windows?"One man's wage rise is another man's price increase." - Harold Wilson
"Fireproof doesn't mean the fire will never come. It means when the fire comes that you will be able to withstand it." - Michael Simmons
"Man who follows car will be exhausted." - Confucius
|
|
|
|
|
| I've never seen one that wasn't preformatted although IIRC alot are defaulting to NTFS now. That said current linux distros can read/write NTFS so that shouldn't be a problem. 3x12=36
2x12=24
1x12=12
0x12=18
|
|
|
|
|
DavidCrow wrote: Are these external HDDs preformatted
In my experience they are (nowadays) usually preformatted with FAT32.
DavidCrow wrote: does that mean files written by Ubuntu can be read by Windows?
Normally, yes."When did ignorance become a point of view" - Dilbert
|
|
|
|
|
Hi, I have a Toshiba Satellite P25-S5092 Laptop. It uses a Wireless that uses Atheros driver software. This brand of toshiba is 2003 edition and it drivers cannot be found on Toshiba website.
I formatted the system and reinstalled it. I cannot use the wireless driver. Please what do I do? Can anyone help me with the Atheros AR5001 from Atheros Network Communications.
I have tried to update / rollback the driver but I kept on receiving error message - 'Not found'
Pls help.
Thanks.
Eyo Honesty
eyo_eyo_honesty@yahoo.co.uk
|
|
|
|
|
Hello all,
I've receivede a call from a friend of mine (and yes, it is a friend it's not me  ) that had installed a software (I don't know which one) and that after restarting the computer a message appeared into the screen but nothing was possible to be done.
After that, he has restarted the machine as it seemed locked and now nothing happens...
By nothing happens I mean that the keyboard led's are not being lid on the starting up of the computer. Neither the display is showing anything...
I guess that that downloaded software was malicious and that has affected somehow the BIOS settings...
1. Is that possible?
2. what would you suggest to try?
THANK YOU. ![Rose | [Rose]](https://codeproject.freetls.fastly.net/script/Forums/Images/rose.gif)
|
|
|
|
|
Joan Murt wrote: 1. Is that possible?
Is it possible that the software killed the hardware? No.
Can software mess up the BIOS settings? Sure. It's just another device with memory in the machine.
If it is the BIOS, I'd say dig around in the motherboard manual and look for the procedure to reset the BIOS to factory defaults. After that, you should be able to re-setup the BIOS and restart the machine.
|
|
|
|
|
Hi all,
I am developing an application that uses SNMP to draw a topological view of a network (currently only layer 3). However, I have hit a very hard stone wall and i fear the answer may be so obvious I am overlooking it.
Basically, how can i find the IP address of a router interface that is directly connected to another router? (I am using cisco 2600 series in my lab) My initial idea was to view the routing table to get a next hop address. This is fine if the routing table contains a next hop address. But, i have found that in many instances the routing table just says "Go out Serial 0/0". It is as if i need to send a probe packet out the interface to say "Hi, who is connected to me and what is your IP"
Here is a scenario of what I am trying to do. I have 2 routers, router A and router B. A work station is running my application on fa0/0 of router A. the two routers are connected using their serial 0/0 interfaces. My app jumps to router A (the workstations DFG) and gathers all it needs. It now needs to jump to router B to gather info about that device. but the routing table on router A contains the exit interface, NOT the next hop address. So how would i find the address of router B's S0/0 interface to allow my app toconnect to it.
Any help is greatly appreciated and let me know if you need a clearer explanation
Kind Regards,
|
|
|
|
|
Not familiar with the Cisco 2600 MIB, (though I'm certain that the info is in there somewhere  ).
An alternative is to send an echo request to a destination beyond the first router (like traceroute does: http://en.wikipedia.org/wiki/Traceroute[^]
Note that the second router can be configured to not answer echo requests, in which case you're still in the dark.
What you don't say of course, is what protocol the routers are running on their S0/0 interface. It's probably not IP, bit some sort of frame-relay, which means that the info will be meaningless on level 3. The routers will map IP routing on top of frame-relay, through a virtual interface, which will contain the IP address.
Some more details on the exact setup would be welcome...
|
|
|
|
|
Hi There Michel,
Thanks for your response, i was hoping i could gather some interest just for a discussion.
I have virtually solved the problem and have a nice working piece of software. Basically i turned to neighbor discovery protocols. In this Case The Cisco Discovery Protocol (CDP). There is a CDP MIB that provides me with a lot of information about what devices are connected to each interface and therefore, I am able to hop between devices because i know connected IP's.
Basically, my prgram jumps to the device the user specifies as the start IP, entering in the SNMP information and the network address they wish to map. My program then goes to that device, gathers all system information about the device, then gathers information about each local interface. Then my program looks into CDP MIB and looks at the CDP Cache (where neighbor details are stored), creates an entry in the database for each connected IP address, then assigns the connected interface entry to the local interface entry. My program then hops to the next device and does the same, so on and so forth. I have parameters in my code to make sure all IPs visited are noted, and a list of IPs to visit to prevent the oprogram from needlessly visiting the same device twice. This way i can track the hop count and limit it if required. At the end of the data collection, my program then analyses the data and draws a network map and listens for traps its a bit like a very basic network management system (extremley basic) but it works sweetly
Obviously this is not a perfect solution because CDP only operates on cisco devices and in some cases, CDP is disabled. There does exist a little protocol called the Link Layer Discovery Protocol (LLDP, refered to as 802.1ab i think). This is a vendor neutral neighbor discovery protocol and operates very much like CDP, but on any vendors device (Cisco, HP etc) providing the device supports it (my Catalyst 2950s do not, but the 2960's do). However, due to my limited budget the equipment i have does not support LLDP *sigh*
Since doing this for my uni project i have become particulary interested in network mapping and management. It really is quite a variedand detailed topic
Comments and thoughts are welcomed
|
|
|
|
|
Hi
I have bought raid controller Intel SASWT4I and computer with motherboard Gigabyte GA-EQ45M-S2.
I have a problem with installation windows xp professional. I installed controller to machine and created array. Array is ok. Now I'm trying to install windows. In bios controller I set boot option to OS only. Windows started installation, formated disk, copyied files and restarted after 15 sec. After restart windows again started installation from cd rom and want to format my disk.
What I have to do to continue installation of windows?
I've changed the boot option in bios controller to bios & os but result is the same.
ThanksI
|
|
|
|
General 
News 
Suggestion 
Question 
Bug 
Answer 
Joke 
Praise 
Rant 
Admin 
Submit an article or tip
