I was afraid of that answer. I feel like sending the whole object to a byte[] would cost just as much if not more time then it currently takes.
On Wednesday, February 26, 2014 10:22:20 PM UTC-5, Ilia Mirkin wrote: > > On Wed, Feb 26, 2014 at 10:13 PM, jCoder <[email protected] <javascript:>> > wrote: > > I was able to change the BufferedImage to an ImageIcon which is > > Serializable. > > This along with using the FST Serialization class/jar I was able to > improve > > the over all write time to about 10 seconds for a total write size of > about > > 140 MB > > Which is still only 14MB/s write speed when I know I can write at > 150+MB/s > > > > My question is can ProtoBuf handle serializing an ImageIcon, Color, or > any > > Java objects besides basic Strings and Ints? and is it faster the FST? > > Protobuf does not handle serialization of Java objects. You can give > it a byte[] array (or an int/float/etc or a String). Hence my > suggestion for the benchmark. I suspect that the majority of the time > is going to computing the byte[] array representing your object, which > you'd have to do for protobuf anyways. > > > > > Also one of my custom object classes has a nested object class inside of > it, > > will this slow down serialization? > > > > > > On Wednesday, February 26, 2014 9:10:24 PM UTC-5, Ilia Mirkin wrote: > >> > >> On Tue, Feb 25, 2014 at 3:03 PM, jCoder <[email protected]> wrote: > >> > I was wondering if anyone had any success implementing this with a > >> > HashMap > >> > that has a pointer to a BufferedImage. > >> > Example: > >> > Map thumb = new HashMap<String, BufferedImage>(); > >> > > >> > I am currently using Serialization with a custom writeObject() and > >> > readObject() to turn the BufferedImage into a byte[] and back again. > >> > > >> > However this process takes roughly 25 second+ for approximately > 18,000 > >> > BufferedImages (size: 16pixels x 12pixels) previously loaded into > memory > >> > to > >> > be serialized into a file (resulting size 9,408 KB). > >> > Please note this all happening on a SSD (so disk write speed should > not > >> > be > >> > an issue). > >> > > >> > There has be a faster way to do this perhaps Protocol Buffers can > help, > >> > I am > >> > just not sure the best way to handle a BufferedImage with it would > be. > >> > > >> > Any help would be greatly appreciated. > >> > >> Try writing a benchmark that simply converts the BufferedImages to > >> byte[] and throws away the results. That's a lower-bound on your > >> overall serialization speed (without switching away from BufferedImage > >> to something else). > >> > >> -ilia > > > > -- > > You received this message because you are subscribed to the Google > Groups > > "Protocol Buffers" group. > > To unsubscribe from this group and stop receiving emails from it, send > an > > email to [email protected] <javascript:>. > > To post to this group, send email to [email protected]<javascript:>. > > > Visit this group at http://groups.google.com/group/protobuf. > > For more options, visit https://groups.google.com/groups/opt_out. > -- You received this message because you are subscribed to the Google Groups "Protocol Buffers" group. To unsubscribe from this group and stop receiving emails from it, send an email to [email protected]. To post to this group, send email to [email protected]. Visit this group at http://groups.google.com/group/protobuf. For more options, visit https://groups.google.com/groups/opt_out.
