Benchmarks for JPEG2000 encoders on CPU and GPU
Below we provide the benchmarks for Fastvideo JPEG2000 Encoder in comparison with other freely available open source J2K encoding solutions. Some of them are CPU-only, while the others use GPU to accelerate JPEG2000 computations.
Approaches for JPEG2000 performance measurements
There are two standard approaches to performance measurements of JPEG2000 codecs, which utilize GPU. They correspond to the two most common use cases for J2K encoders and decoders.
1. Single image mode consists in processing of single image and could be called "latency-oriented" or "minimum latency" approach. In that case the time interval (latency) between availability of original image in RAM and availability of the processed image in RAM is measured. It means that software cannot expect that any additional images will be processed at the same time and therefore cannot take advantage of multiple image encoding or decoding. Overlapping of current image processing with other activities is undesirable because it would increase the delay for getting the result. We need single image mode almost in all camera applications because apart from JPEG2000 encoding we also have to implement other image processing algorithms at the same pipeline. You can get more info from our Image & Video Processing SDK.
2. Batch mode consists in processing of batch of images and could be called "throughput-oriented" or "maximum performance". In that case frame rate becomes more important feature. It is calculated via division of the total time of processing by the number of processed images. Some JPEG2000 codecs are optimized for this use case, meaning that exploiting of task parallelism leads to better frame rate (throughput) at the expense of increased processing time for separate images. It is possible, because we actually have three devices (CPU, GPU and bus interface between them), which can be used simultaneously in that mode, whereas at single image mode these devices are used sequentially for different stages of JPEG2000 algorithm. Moreover, GPU can process several images simultaneously to increase frame rate even more, if each image is too small to load a multitude of GPU cores (especially at Tier-1 stage). Important limitation for simultaneous processing of several images is imposed by amount of free GPU memory. Batch mode is a must for streaming applications when the pipeline consists of JPEG2000 encoder or decoder. For more complicated workflow it's better to utilize single image mode, though the performance will be less.
Briefly, JPEG2000 batch mode can take into account specific methods of task parallelism, based on the following:
Both the above modes (single image and batch) can not be fully applicable to CPU-based JPEG2000 solutions because in such cases everything is done on CPU. That's why we can consider multithreaded CPU-based JPEG2000 encoding/decoding to be alike single image mode.
At the moment we don't consider here the following possible modes for JPEG 2000 benchmarking on GPU:
Results for all modes will be published as soon as their implementations are ready.
We don't hide anything concerning benchmarking procedures and achieved results. That's why all our users can always reproduce our benchmarks because we publish not only timing and performance - we supply full info about hardware, JPEG2000 parameters, test images and testing modes.
JPEG 2000 encoding benchmarks
We've carried out time and performance measurements for JPEG2000 encoding for 24-bit images with 2K and 4K resolutions. All results don't include any host I/O latency (image loading to RAM from HDD/SSD and saving back) and we've also excluded host-to-device transfer time. We've done such an assumption to reproduce J2K encoder usage in our conventional image processing pipeline, when initial data reside in GPU memory. Results for GPU-based JPEG2000 encoding software also include Tier-2 time on CPU, because this stage in our implementation is performed on CPU. In the tables below one can find averaged results for the best series of 1000 measurements.
JPEG2000 encoding parameters
Hardware and software
JPEG2000 Encoders for comparison
JPEG2000 lossy encoding at single image mode for 2K image: 2k_wild.ppm (1920×1080, 4:4:4, 24-bit)
JPEG2000 lossy encoding at single image mode for 4K image: 4k_wild.ppm (3840×2160, 4:4:4, 24-bit)
MB/s – MegaBytes per second
Fig.1: Fastvideo JPEG2000 performance on GeForce GTX 1080 (lossy encoding, single image mode)
From the above figure we can see the encoding speed (JPEG 2000 performance for lossy compression) as a function of image size for Fastvideo JPEG2000 encoder at single image mode. Maximum JPEG2000 performance could be achieved with codeblock size 32×32 in most cases. For images with frame size more than 6 MB, preferred codeblock size is 32×32 at single image mode. It could also be seen that there is a performance saturation, which is dependent on image size for different codeblocks. This is a key point to get better results at batch mode. For 8K image compression with visually lossess parameters, performance saturation is present for any codeblock size at single image mode.
Figure 1 shows that on NVIDIA GeForce GTX 1080 it's possible to achieve important milestones at single image mode for visually lossless JPEG2000 encoding. For codeblocks 16×16 one can overcome 900 MB/s performance, for codeblocks 32×32 maximum performance exceeds 1300 MB/s, for codeblocks 64×64 maximum performance could reach 1100 MB/s. Performance saturation for codeblocks 16×16 occurs at 4K resolution for vusually lossless compression.
Fig.2: Fastvideo J2K performance as a function of compression ratio (lossy encoding, single image mode)
Figure 2 shows Fastvideo JPEG 2000 encoder performance as a function of compression ratio for different image resolutions for lossy compression at single image mode with standard testing conditions as stated above.
Lossless JPEG2000 encoding at single image mode for 2K image: 2k_wild.ppm (1920×1080, 4:4:4, 24-bit)
Lossless JPEG2000 encoding at single image mode for 4K image: 4k_wild.ppm (3840×2160, 4:4:4, 24-bit)
Superior performance of JPEG 2000 encoding at batch multithreaded mode
For batch mode we've carried out performance measurements for JPEG 2000 encoding exactly with the same parameters as we used at single image mode. All results don't include host I/O latency (image loading to RAM from HDD/SSD and saving back).
To get maximum performance at batch mode, we don't need very large images as for single image mode. For example, 4K image contains 4 times more pixels compared to 2K. It means that at batch mode we can expect that encoding time for 2K will be 4 times less than for 4K. In theory, if at single image mode we can do visually lossless JPEG2000 encoding for 2K at 140 fps and for 4K at 52 fps (Fig.1, codeblock size 32×32), then it could be possible to achieve frame rate 52*4=208 fps for 2K encoding speed by processing 4 images with 2K resolution simultaneously, as a batch. We could expect even higher speedup for lossy compression, if at single image mode GPU is not completely occupied with 4K images (see Fig. 1) and batch size for 2K is more than 4. If we also take into account simultaneous processing on both CPU and GPU, which is possible at batch mode, one could get additional acceleration.
Fastvideo JPEG2000 lossy encoding benchmarks at batch mode
Fastvideo JPEG2000 lossless encoding benchmarks at batch mode
To the best of our knowledge, the above J2K performance benchmarks for lossy and lossless encoding are the fastest among all existing open source and commercial JPEG2000 encoders on CPU or GPU both for single image mode and for batch mode. To make it transparent and simple, we have published all info concerning time measurements, together with sample images, JPEG2000 parameters and hardware specifications to offer everyone an opportunity to reproduce our results and to check performance measurements of other J2K encoders at the same testing conditions. Demo for Windows for our J2K encoder on GPU could be downloaded here.
Please let us know about your performance results for software JPEG2000 encoders that you could have: Aware, Comprimato, Elecard, ERDAS ECW, FFmpeg, Kakadu, Leadtools, Lizardtech, Lurawave, Mainconcept, Morgan, etc.