Does SSD improve performance?

Got a small 250GB SSD drive, USB 3.0. I use it for Virtualization and some IT stuff, has some important files on it. So before formatting or give it a try, I wonder if someone is currently using one and had the chance to compare to a regular drive.

I can still move the file to another drive or PC, but I would prefer to avoid all that hassle and waste my time if the speed won’t make a huge difference in user’s experience.

@sql_lover - Nope, there’s no advantage to having a faster drive so keep your SSD for your other tasks.

Just so you know, I believe there is little to no advantage because of the available bandwidth of the USB ports.  That will be your bottleneck.

@mrv77 is right. But we don’t even need the speed of a USB 3 port which is why we picked USB 2 vs. 3. 


So, Tablo USB ports are USB 2.0? If that’s the case then that’s my answer there.

Yep - they’re USB 2. 

Did you try this ? It seems like the initial playback delay would be better since the drive does not need to spin up… otherwise no difference.

The only advantage to SSD I can see is because they are not a spinning media, if for example you accidentally knock over the drive while its recording {cough cough cough} and it falls on its side and hits pretty hard you are far less likely to damage the drive or even interrupt the recording in any significant way.

An SSD is also going to be quiet. I have one of the bigger self-powered external drives, and the sound it makes is noticeable at times. The main reason not to use an SSD is price. It’s not going to significantly improve Tablo’s performance, while it will cost a lot more than a normal external drive.


For best performance, I recommend a drive with a random read and write speed of at least 60MBps. That way, the drive won’t be a bottleneck in any way.

60MBps is 480Mbps which is the max bandwidth of a USB 2.0 connection (which is what the Tablo has). With overhead, there’s no way a drive on a USB 2.0 connection would even come close to what you’re recommending.


So, then you agree with what I said, the drive won’t be a bottleneck in any way.

I guess, but try to be realistic about what you’re recommending as an “at least” speed. I mean, I could say, for super premium best performance and to eliminate all bottlenecks it is best to have a RAID 10 drive array attached to the a Tablo which will at least have 450MBps read write throughout. But, would that be realistic?

More realistic would be: The max single stream bandwidth for a Tablo is 10Mbps (the 1080p or 720p @ 60fps setting). The Tablo can serve out video to six clients and record up to four streams. That’s a possible ten streams. 10 X 10Mbps is 100Mbps max. 100Mbps is 12.5MBps. The maximum theoretical throughout of a Tablo is 1/5 of what you’re recommending as a minimum throughout on a drive to eliminate bottlenecks.

For me a recommended “at least” speed is the minimum speed for a system to work properly, not the theoretical max of the bus that a drive is connected to. So, given the max bandwidth that could theoretically be used by a Tablo is 12.5MBps, yea 60MBps is nice, but definitely not what is needed to run the system without bottlenecking it.

actually, the strength of an SSD is access time due to a lack of the access delay of a spinning disk. This is why running a SSD as the primary drive makes the entire system “feel” more responsive. (HDD: around 9 milliseconds, SSD around .02 milliseconds)

Running a SSD as the storage drive for a Tablo would improve performance, but that improvement would be undetectable to any real person when compared to a reasonably fast spinning hard drive. In order to see the difference you would need a lot more simultaneous clients than what the Tablo supports.


I stand corrected. It’s good to see someone do the math. I had a WD My Passport Enterprise 500GB USB 3.0/2.0 HDD that got about 40MB/s read/write, but it would buffer when when there were combinations of playing or viewing on three or four Roku devices. I switched it with a SanDisk Extreme Pro 64MB USB 3.0/2.0 (Read 245MB/s, Write 190MB/s) flash drive and I have no issues now. Perhaps someone can give me an idea why the WD My Passport Enterprise didn’t do as good as the SanDisk Extreme Pro if all I needed was 12.5MB/s.

The throughput is just one metric to “judge” a drive by. I think if you’re accessing the Tablo from multiple devices, the issue is caused by the mechanical nature of the drive. The WD My Passport is a mechanical drive so the drive arm was probably being driven nuts by seeks and reads based all the activity on the drive. iops (Input Output Per Second) is what this metric is. Mechanical drives have much lower iops than SSD drives since SSD drives do not have anything that is reliant on mechanical operations. The difference is in orders of magnitude – Mechanical drives are in the 200 iops or below. SSD drives tend to be in the 1000s of iops.

Wouldn’t a random read and write of at least 12.5MB/s be enough regardless whether it’s a mechanical HDD or a SSD? A random sustained throughput of 12.5MB/s is just that, 12.5MB/s no matter what the seek latency is except for perhaps the initial seek. I thought that random read/write performance takes into account the seek time.


My Tablo has an SSD. Its super quiet and seeking/skipping around in shows is near instant.

Just brainstorming here…

Wouldn’t a SSD use too much power for the Tablo to handle?
Or at least, cause it to run too hot, or something like that?