I read in the user manuals:
Lime-A10: draws between 400mA and 650mA/750mA (inconsistency in the PDF)
Lime2-A20: draws between 400mA and 650mA
(from 5V, without external peripherals)
in the introduction sections of some of the manuals (and possible in some parts of the website),
it is stated that the A10 is a less powers-hungry alternative to the A20.
So the consumption mentioned and this later statement somehow seem to contradict...
or can you please explain?
Im considering whether to buy Lime-A10 or Lime2-A20 for a linux server, and power consumption is the main argument for me, while second is CPU & RAM. I am most interested in idle state power consumption.
i might run stuff like tor and jitsi relays. it will be a headless system (no graphics).
Do you want min / max / typ figures? You look to have max ones only.
Realistically you should just measure with your own workload but a 2-core is likely to use a bit more than a single core.
The boards are very cheap so maybe just buy both and then measure or start with A10 and see if it's OK.
John
If the load is the same, the power consumption should not make a big difference. The A20 will calculate shorter with 2 cores and a higher peak consumption during this time, the A10 will calculate longer with lower peak consumption. In the end almost the same Wh will be used.
In idle state the power consumption is almost the same, the lime2 will use some 10mA more because of more RAM and Gbit ethernet, but the cores in idle state differ only by some few mA.
If i would have to choose i would always prefer a dual core because of the better response times and more cpu power reserves. If you have a "hanging" process with 100% CPU you still have 1 core left for your ssh session.
More important is to take a really fast SD-Card so the system does not wait-freeze on HD-IO. For reliability take a industry-grade card like the SDSDSQAE-032G or similar. For a server system try to avoid consumer cards like Extreme/Extreme pro.
And adding this:
The A10 is made in 55nm, the A20 in 40nm, so the A20 has the process advantage concerning power consumption.
wow, thanks a lot for the info. all i wished for and more. :-)
now that you made me think about the storage, soenke...
* what are the specs for the microSD cards that olimex sells (with android/debian)? (size in GB and read&write speeds)
* does it make sense to get a microSD card and a USB flash drive of the same size, and set them up as a RAID0?
* is the Lime2 USB v 2.0 or 3.0? the features list just states "high-speed"
lime2 has USB2.0
I dont know how fast the olimex cards are, i guess they are standard-class10 (>10MB/s) and for sure not the fastest on the market. And that for a good reason, their pre-made cards are for people who start playing around with embedded boards and linux. So it would be a waste of money to use big, fast and expensive cards for that.
If you compare them to industry grade cards, the latter have higher transfer speeds, faster random access (which is the most important for running an os) and of course a higher long-term reliability.
But for just playing around, principal tests and development the olimex-cards are by far good enough.
The cards i use (mentioned above) have 32GB, transfer rates of about 65MB/s and faster random access speed than an extreme pro but cost about 20€ per card.
RAID0 rarely makes sense on flash devices, especially with different interfaces. I would not do that.
Better invest your money in a bigger and/or faster microSD card. If that is not enough, there is a SATA-connector onboard...
again, thanks a lot soenke, very good info!
i don't think i ever received that good help online, in the last 15 years.
i salute you!
the only thing i am left with... what would you suggest for minimizing the risk of loosing data due to hardware failure, using lime2 as a server, and trying to consume little space and power?
using an SD card for OS and data storage, and a SATA or USB drive, or simply a USB flash stick as a backup storage?
you are welcome, glad i can help :)
simple answer: backup!
If it is really vital data, it shall not be physically connected to the server for longer than the backup runs or have another backup in spare (e.g. another medium from the previous day). Depending on your requirements you can e.g. exchange the backup medium every day or every week cycling with 2-7 different backup media.
Another solution is another computer running in another location, like a cloud backup, a root-server or another lime2 at your parents home.
If 32GB or 64GB is enough for your system + data i would only use a single SD Card for everything, that is the smallest and most power saving configuration.
ouhhh really? so it is unlikely that an SD card is going to fail completely and more likely that just certain "blocks" are going to fail? otherwise backup on the same card would not make sense, right?
i plan to be mostly away from the server(s), and would prefer not having to exchange backup media. and yes, i plan on having little data overall.
i was actually thinking about doing exactly that" a second lime2 at my parents home. :D
.. at least for the second/backup machine, wake-on-LAN would be a good thing i guess. is that possible with the built-in ethernet?
In my experience SD cards fail blockwise before they die completely. Like a SSD they can reassign some reserve space if single cells fail, you will _not_ get a hint if that happens. The controller also mangages wear-leveling, so avoid having a 90% full card and lots of write cycles in the last 10%. But if the controller on the card dies, the whole card is dead instantly. Sometimes you will get hints by kernel freezes that something is dying but not all the time.
A backup on the same media is not a backup :) That can still save you from accidental deletion of some files but this is not called a "real" backup protecting against hardware failure, lightning strikes etc.
I have not done anything with WOL yet so i cant tell you that. Maybe searching for the PHY in the internet helps. I think it depends on the linux driver if it is supported.
RTL8211CL-GR is the PHY chip on lime2, and it seems not to support WOL (datasheet):
http://www.datasheet4u.com/datasheet/R/T/L/RTL8211C-GR_RealtekMicroelectronics.pdf.html
also, the BIOS would have to support it, if i get it right. i don't know anything about the lime2 BIOS, but i read that most BIOSes do support it (i guess there it is mostly a software/firmware issue).
either way, i will assume this is not supported, and that is ok for me, no absolute must have, but...
as i read that they are, or might be looking for a new PHY that would support industry-level temperatures, it would be nice if that new PHY would also support WOL (hope someone is listening here ;-) ), as it is useful for a low-load server setup.
thanks again, a lot, soenke! :-)
hey soenke! :-)
i try to choose an SD car now, and looking at what you posted back then, i am quite confused.
Quote from: soenke on May 09, 2016, 12:19:39 PM
More important is to take a really fast SD-Card so the system does not wait-freeze on HD-IO. For reliability take a industry-grade card like the SDSDSQAE-032G or similar. For a server system try to avoid consumer cards like Extreme/Extreme pro.
reading about SanDisk's Ultra and Extreme lines, supposedly Ultra is for consumers, and Extreme/Extreme Pro is for professionals, and none is explicitly industry-grade. i assume the one you thouhg of is ultra, which supposedly is the cheap version(?) also, i think you had at least one typo in the model you mentioned, as a google search for it turns out only this forum post.
is it sufficient to say, that we want at least 48MB/s read speed? if so, 32GB can be found for 10Eur na d64 for 20Eur (September 2016).
Get a Samsung EVO 32/64
http://forum.armbian.com/index.php/topic/954-sd-card-performance/
Hi hoijui,
i did not mistype the sd-card manufacturer number, it is the one we use in our boards.
The reason why we use this one is, that these are of higher quality and have a better random read/write performance than the exteme pro, but also a slower sequential read/write performance. The latter does not matter as it is limited by the board anyways, as you can read in igors link.
The extreme pro is optimized for video cameras or professional photo cameras which have to write a lot (sequentially) in a short time. It is not optimized for running operating systems.
Our experience with the extreme pro (and other consumer cards of that class) is, that they tend to randomly fail after a while or cause random kernel freezes. Not all of them, but about 20% (we used about 100 of them, about 20 failed or caused crashes after 3 hours to 6 month).
Since we the use industry grade sd-cards (SDSDSQAE with 32GB) we have had not a single failure or crash so far.
You can order them at your local sandisk distributor. afaik you cant buy them at consumer shops like amazon, because there they sell the cards which are not good enough for the industry as exteme/extreme pro ;)
despite looking in many places, and writing emails to local distributors, and looking at farnell's website, i could not yet find a single trace of the ominous SDSDSQAE cards mentioned (am in germany here).
what i was able to find, is "industrial grade" SD and MicroSD cards. there are some industrial grade SD cards with sizes up to max 4GB (not enough for a server OS + data, for me), which already cost 80+Eur, and then some industrial grade panasonic card sup to 32GB for ~170Eur, for which i could build a RAID5 out of 4 1TB SATA drives. i also found 32GB kingston "industrial (temperature)" microSD cards for around 25Eur, which would be fine, but i kind of assume that this "industrial" really just refers to the temperature range, and does not imply any improved reliability (at normal temperatures)... am i right there?
if yes, then i see no other option then to go SATA, which is a real pity, because of weight, physical size, noise and power consumption.
did i miss something, or get something wrong?
i also found this info:
https://www.sandisk.com/oem-design/industrial/industrial-cards
but no way to buy those anywhere, still.
Quote from: hoijui on December 09, 2016, 02:40:44 PM
despite looking in many places, and writing emails to local distributors, and looking at farnell's website, i could not yet find a single trace of the ominous SDSDSQAE cards mentioned (am in germany here).
what i was able to find, is "industrial grade" SD and MicroSD cards. there are some industrial grade SD cards with sizes up to max 4GB (not enough for a server OS + data, for me), which already cost 80+Eur, and then some industrial grade panasonic card sup to 32GB for ~170Eur, for which i could build a RAID5 out of 4 1TB SATA drives. i also found 32GB kingston "industrial (temperature)" microSD cards for around 25Eur, which would be fine, but i kind of assume that this "industrial" really just refers to the temperature range, and does not imply any improved reliability (at normal temperatures)... am i right there?
if yes, then i see no other option then to go SATA, which is a real pity, because of weight, physical size, noise and power consumption.
did i miss something, or get something wrong?
A lot of industrial grade SD cards are exactly the same as normal ones, except they are more expensive. SD media is critical element by default, no matter where you will use it. Just go for Samsung EVO and give 100EUR to charity instead. There is another general problem regarding SD cards ... there are so many fakes out there that you need a luck to get genuine one in first place. Even from respected shops.
SATA. If you are thinking of RAID with something with 25 EUR tag:
(http://img.dxcdn.com/productimages/sku_370801_1.jpg)
than again, forget about, since it's unreliable and RAID won't help protect your data.
i bought them at www.active-components.eu (i am also from germany).
I phoned with some sales guy from sandisk directly and he explained me that there is a significant difference between consumer-grade sandisk products like extreme pro and those industrial-grade cards. He said that they are more reliable and more optimized for random-access than the extreme pro. That goes with a reduced max. seq. transfer rate.
QuoteI phoned with some sales guy from sandisk directly
Well, sales people surely knows how things are ;D
BTW: We did a lot of tests with different SD cards and we were looking exactly for best "random-access":
https://forum.armbian.com/index.php/topic/954-sd-card-performance
Most expensive AKA PRO cards were not doing anything better, some of them were actually worse.
Add: you are welcome to add the same tests with those "industrial grade" cards.
Our fastest measurement was 8GB eMMC via SD, which costs 30 EUR @Pollin, Germany and I haven't saw anything close to this:
random random
kB reclen write rewrite read reread read write
102400 4 11485 12921 7044 7060 7010 11957
102400 16 17668 19117 17584 17579 17467 16706
102400 512 22711 22641 22962 22967 22942 21219
102400 1024 22800 22871 23056 23054 23041 22737
102400 16384 22803 22852 22990 22989 22989 22764
This is my measurement of the SDSDSQAE with 32GB in one of our test systems.
Olimex A10S, Allwinner Kernel 3.4, ext4, nojournal, about 1 year old running linux 24/7 and a lot i/o stuff like measurement data aquisition etc.
Command line used: iozone -e -I -a -s 100M -r 4k -r 16k -r 512k -r 1024k -r 16384k -i 0 -i 1 -i 2
Output is in Kbytes/sec
Time Resolution = 0.000001 seconds.
Processor cache size set to 1024 Kbytes.
Processor cache line size set to 32 bytes.
File stride size set to 17 * record size.
random random
KB reclen write rewrite read reread read write
102400 4 2697 2817 8132 8074 8031 2674
102400 16 4770 5232 14553 14581 14055 6041
102400 512 16128 17570 19280 19274 19271 13255
102400 1024 16198 17584 19297 19294 19285 15468
102400 16384 17575 17713 19429 19505 19468 17440
Still, the main reason we chose this card is reliability. We had about 10% failing extreme pro cards before we switched to this card. So far not a single card (of some 100s) failed in the last year.
QuoteStill, the main reason we chose this card is reliability. We had about 10% failing extreme pro cards before we switched to this card. So far not a single card (of some 100s) failed in the last year.
Reliability might be accountable, but is something hard to argue about. We don't have enough data to make any conclusions. I don't like to speculate like this: extreme pro cards were made for fast video recording, for cameras. For such usage they might not fail so soon, while they do on lot's of random write which is normal in SBC world ...
Proper reliability - "burn in" test across wide range of card is too big cost for us, so we will probably never do it unless someone puts cash on the table to cover related costs (https://forum.armbian.com/index.php/topic/954-sd-card-performance/?p=19932). Our small test was already on the edge of the hobby project, with cards which we had around. On top of this no one wanted to destroy them on the test :)
I think it is more like a selection process and some changes in parameters of the cards controller. I guess they use the upper 10% of NAND chips and controllers for more reliable cards and do a more conservative parameterization. That might also be the reason why the max. seq. transfer rate is slower.
The second somewhat 20% selection they might use in enterprise SSD, then maybe 40% for consumer SSDs. After that maybe 10% for consumer sd-cards, and the crappy rest of 10% is integrated into 3rd-party-USB-Sticks ;)
Still, the most important value of iozone for us is the 4k random read and this card is one of the fastest in this. Yes, it would be nice if random write would be faster like on a emmc, but we gladly trade this for the increased reliability and 0% dead-on-arrival-rate. So if someone has his priority on a rock-stable sd-card, chose this one.
And for the cost: it has almost the same price (even 1-2€ cheaper) as the extreme pro.