WEBVTT
Kind: captions
Language: en

00:00:12.889 --> 00:00:16.820 align:start position:0%
 
my<00:00:13.889><c> name</c><00:00:14.039><c> is</c><00:00:14.160><c> Chicago</c><00:00:14.699><c> I'm</c><00:00:15.209><c> I</c><00:00:15.869><c> work</c><00:00:16.320><c> for</c><00:00:16.529><c> stream</c>

00:00:16.820 --> 00:00:16.830 align:start position:0%
my name is Chicago I'm I work for stream
 

00:00:16.830 --> 00:00:19.340 align:start position:0%
my name is Chicago I'm I work for stream
Bo<00:00:16.980><c> I'm</c><00:00:17.250><c> one</c><00:00:18.090><c> of</c><00:00:18.240><c> the</c><00:00:18.359><c> cofounders</c><00:00:18.539><c> upstream</c>

00:00:19.340 --> 00:00:19.350 align:start position:0%
Bo I'm one of the cofounders upstream
 

00:00:19.350 --> 00:00:22.580 align:start position:0%
Bo I'm one of the cofounders upstream
neo<00:00:19.530><c> and</c><00:00:19.710><c> in</c><00:00:20.520><c> extremely</c><00:00:21.270><c> or</c><00:00:21.360><c> I</c><00:00:21.390><c> mostly</c><00:00:22.320><c> focus</c>

00:00:22.580 --> 00:00:22.590 align:start position:0%
neo and in extremely or I mostly focus
 

00:00:22.590 --> 00:00:25.910 align:start position:0%
neo and in extremely or I mostly focus
on<00:00:22.980><c> mastering</c><00:00:23.400><c> in</c><00:00:23.580><c> star</c><00:00:23.850><c> H</c><00:00:24.030><c> pieces</c><00:00:24.450><c> and</c><00:00:24.920><c> before</c>

00:00:25.910 --> 00:00:25.920 align:start position:0%
on mastering in star H pieces and before
 

00:00:25.920 --> 00:00:28.400 align:start position:0%
on mastering in star H pieces and before
stream<00:00:26.820><c> deal</c><00:00:26.970><c> I</c><00:00:27.210><c> was</c><00:00:27.360><c> their</c><00:00:27.510><c> Twitter</c><00:00:27.750><c> leading</c>

00:00:28.400 --> 00:00:28.410 align:start position:0%
stream deal I was their Twitter leading
 

00:00:28.410 --> 00:00:38.240 align:start position:0%
stream deal I was their Twitter leading
the<00:00:28.590><c> messaging</c><00:00:29.220><c> infrastructure</c><00:00:36.470><c> so</c><00:00:37.470><c> the</c><00:00:38.040><c> key</c>

00:00:38.240 --> 00:00:38.250 align:start position:0%
the messaging infrastructure so the key
 

00:00:38.250 --> 00:00:41.680 align:start position:0%
the messaging infrastructure so the key
that<00:00:39.030><c> key</c><00:00:39.239><c> problem</c><00:00:39.690><c> are</c><00:00:40.010><c> able</c><00:00:41.010><c> to</c><00:00:41.040><c> do</c><00:00:41.190><c> it</c>

00:00:41.680 --> 00:00:41.690 align:start position:0%
that key problem are able to do it
 

00:00:41.690 --> 00:00:44.870 align:start position:0%
that key problem are able to do it
so<00:00:42.690><c> the</c><00:00:43.050><c> key</c><00:00:43.590><c> engineering</c><00:00:44.130><c> problem</c><00:00:44.579><c> for</c>

00:00:44.870 --> 00:00:44.880 align:start position:0%
so the key engineering problem for
 

00:00:44.880 --> 00:00:47.150 align:start position:0%
so the key engineering problem for
messaging<00:00:45.510><c> and</c><00:00:45.630><c> storage</c><00:00:46.050><c> extremely</c><00:00:46.920><c> are</c><00:00:47.010><c> we</c>

00:00:47.150 --> 00:00:47.160 align:start position:0%
messaging and storage extremely are we
 

00:00:47.160 --> 00:00:50.950 align:start position:0%
messaging and storage extremely are we
are<00:00:47.250><c> facing</c><00:00:47.430><c> is</c><00:00:48.530><c> mostly</c><00:00:49.530><c> one</c><00:00:50.100><c> is</c><00:00:50.280><c> the</c><00:00:50.460><c> scale</c>

00:00:50.950 --> 00:00:50.960 align:start position:0%
are facing is mostly one is the scale
 

00:00:50.960 --> 00:00:53.270 align:start position:0%
are facing is mostly one is the scale
scalability<00:00:51.960><c> issue</c><00:00:52.140><c> for</c><00:00:52.470><c> messaging</c><00:00:52.890><c> system</c>

00:00:53.270 --> 00:00:53.280 align:start position:0%
scalability issue for messaging system
 

00:00:53.280 --> 00:00:57.139 align:start position:0%
scalability issue for messaging system
and<00:00:53.460><c> also</c><00:00:54.030><c> the</c><00:00:54.360><c> simplest</c><00:00:55.070><c> simplicity</c><00:00:56.149><c> the</c>

00:00:57.139 --> 00:00:57.149 align:start position:0%
and also the simplest simplicity the
 

00:00:57.149 --> 00:00:59.150 align:start position:0%
and also the simplest simplicity the
unification<00:00:57.809><c> of</c><00:00:57.960><c> messaging</c><00:00:58.440><c> and</c><00:00:58.469><c> storage</c>

00:00:59.150 --> 00:00:59.160 align:start position:0%
unification of messaging and storage
 

00:00:59.160 --> 00:01:09.499 align:start position:0%
unification of messaging and storage
system<00:01:06.229><c> so</c><00:01:07.229><c> the</c><00:01:07.590><c> scaling</c><00:01:08.069><c> issue</c><00:01:08.280><c> comes</c><00:01:09.090><c> from</c>

00:01:09.499 --> 00:01:09.509 align:start position:0%
system so the scaling issue comes from
 

00:01:09.509 --> 00:01:11.870 align:start position:0%
system so the scaling issue comes from
our<00:01:10.079><c> messaging</c><00:01:10.499><c> system</c><00:01:10.799><c> of</c><00:01:10.950><c> our</c><00:01:11.490><c> storage</c>

00:01:11.870 --> 00:01:11.880 align:start position:0%
our messaging system of our storage
 

00:01:11.880 --> 00:01:15.410 align:start position:0%
our messaging system of our storage
system<00:01:12.270><c> comes</c><00:01:12.539><c> from</c><00:01:12.780><c> two</c><00:01:13.020><c> aspects</c><00:01:13.469><c> is</c><00:01:14.179><c> one</c><00:01:15.179><c> is</c>

00:01:15.410 --> 00:01:15.420 align:start position:0%
system comes from two aspects is one is
 

00:01:15.420 --> 00:01:17.989 align:start position:0%
system comes from two aspects is one is
the<00:01:15.889><c> scalability</c><00:01:16.889><c> of</c><00:01:16.920><c> the</c><00:01:17.130><c> number</c><00:01:17.609><c> of</c><00:01:17.819><c> the</c>

00:01:17.989 --> 00:01:17.999 align:start position:0%
the scalability of the number of the
 

00:01:17.999 --> 00:01:21.349 align:start position:0%
the scalability of the number of the
topics<00:01:18.420><c> in</c><00:01:18.630><c> a</c><00:01:18.749><c> messaging</c><00:01:19.499><c> system</c><00:01:20.029><c> and</c><00:01:21.029><c> the</c>

00:01:21.349 --> 00:01:21.359 align:start position:0%
topics in a messaging system and the
 

00:01:21.359 --> 00:01:23.690 align:start position:0%
topics in a messaging system and the
other<00:01:21.509><c> issue</c><00:01:21.959><c> is</c><00:01:22.139><c> the</c><00:01:22.619><c> scalability</c><00:01:23.009><c> of</c><00:01:23.609><c> the</c>

00:01:23.690 --> 00:01:23.700 align:start position:0%
other issue is the scalability of the
 

00:01:23.700 --> 00:01:27.980 align:start position:0%
other issue is the scalability of the
system<00:01:24.509><c> to</c><00:01:24.719><c> be</c><00:01:24.899><c> able</c><00:01:25.079><c> to</c><00:01:25.289><c> scale</c><00:01:25.649><c> beyond</c><00:01:26.990><c> number</c>

00:01:27.980 --> 00:01:27.990 align:start position:0%
system to be able to scale beyond number
 

00:01:27.990 --> 00:01:29.929 align:start position:0%
system to be able to scale beyond number
of<00:01:28.229><c> hundreds</c><00:01:28.709><c> or</c><00:01:28.979><c> maybe</c><00:01:29.159><c> thousands</c><00:01:29.639><c> of</c><00:01:29.729><c> the</c>

00:01:29.929 --> 00:01:29.939 align:start position:0%
of hundreds or maybe thousands of the
 

00:01:29.939 --> 00:01:31.430 align:start position:0%
of hundreds or maybe thousands of the
nodes

00:01:31.430 --> 00:01:31.440 align:start position:0%
nodes
 

00:01:31.440 --> 00:01:37.410 align:start position:0%
nodes
[Music]

00:01:37.410 --> 00:01:37.420 align:start position:0%
 
 

00:01:37.420 --> 00:01:40.020 align:start position:0%
 
I<00:01:37.450><c> think</c><00:01:37.990><c> that</c><00:01:38.200><c> the</c><00:01:38.860><c> challenging</c><00:01:39.460><c> things</c><00:01:39.820><c> for</c>

00:01:40.020 --> 00:01:40.030 align:start position:0%
I think that the challenging things for
 

00:01:40.030 --> 00:01:44.490 align:start position:0%
I think that the challenging things for
the<00:01:40.659><c> scalability</c><00:01:41.049><c> issue</c><00:01:41.680><c> on</c><00:01:42.189><c> perhaps</c><00:01:43.149><c> on</c><00:01:44.079><c> the</c>

00:01:44.490 --> 00:01:44.500 align:start position:0%
the scalability issue on perhaps on the
 

00:01:44.500 --> 00:01:47.429 align:start position:0%
the scalability issue on perhaps on the
messaging<00:01:44.829><c> system</c><00:01:44.890><c> its</c><00:01:45.670><c> various</c><00:01:46.439><c> especially</c>

00:01:47.429 --> 00:01:47.439 align:start position:0%
messaging system its various especially
 

00:01:47.439 --> 00:01:50.940 align:start position:0%
messaging system its various especially
for<00:01:47.680><c> the</c><00:01:48.159><c> arise</c><00:01:48.840><c> micro</c><00:01:49.840><c> service</c><00:01:50.200><c> you</c><00:01:50.740><c> might</c>

00:01:50.940 --> 00:01:50.950 align:start position:0%
for the arise micro service you might
 

00:01:50.950 --> 00:01:54.870 align:start position:0%
for the arise micro service you might
need<00:01:51.100><c> a</c><00:01:51.369><c> lot</c><00:01:51.640><c> of</c><00:01:52.080><c> small</c><00:01:53.080><c> a</c><00:01:53.590><c> lot</c><00:01:54.189><c> of</c><00:01:54.399><c> small</c>

00:01:54.870 --> 00:01:54.880 align:start position:0%
need a lot of small a lot of small
 

00:01:54.880 --> 00:01:59.460 align:start position:0%
need a lot of small a lot of small
toppings<00:01:55.570><c> or</c><00:01:57.390><c> messaging</c><00:01:58.390><c> cues</c><00:01:58.659><c> for</c><00:01:59.079><c> micro</c>

00:01:59.460 --> 00:01:59.470 align:start position:0%
toppings or messaging cues for micro
 

00:01:59.470 --> 00:02:03.210 align:start position:0%
toppings or messaging cues for micro
service<00:01:59.860><c> and</c><00:02:00.509><c> the</c><00:02:01.509><c> ability</c><00:02:01.960><c> to</c><00:02:01.990><c> scale</c><00:02:02.320><c> that</c><00:02:02.890><c> is</c>

00:02:03.210 --> 00:02:03.220 align:start position:0%
service and the ability to scale that is
 

00:02:03.220 --> 00:02:06.660 align:start position:0%
service and the ability to scale that is
you<00:02:03.640><c> you</c><00:02:04.119><c> need</c><00:02:04.390><c> to</c><00:02:04.509><c> scale</c><00:02:04.750><c> the</c><00:02:05.259><c> i/o</c><00:02:06.100><c> all</c><00:02:06.549><c> the</c>

00:02:06.660 --> 00:02:06.670 align:start position:0%
you you need to scale the i/o all the
 

00:02:06.670 --> 00:02:09.020 align:start position:0%
you you need to scale the i/o all the
data<00:02:06.850><c> the</c><00:02:07.209><c> stupid</c><00:02:07.630><c> but</c><00:02:08.259><c> you</c><00:02:08.380><c> also</c><00:02:08.560><c> need</c><00:02:08.799><c> to</c>

00:02:09.020 --> 00:02:09.030 align:start position:0%
data the stupid but you also need to
 

00:02:09.030 --> 00:02:11.910 align:start position:0%
data the stupid but you also need to
scale<00:02:10.030><c> the</c><00:02:10.390><c> number</c><00:02:10.420><c> the</c><00:02:11.080><c> metadata</c><00:02:11.530><c> and</c>

00:02:11.910 --> 00:02:11.920 align:start position:0%
scale the number the metadata and
 

00:02:11.920 --> 00:02:15.300 align:start position:0%
scale the number the metadata and
scaling<00:02:12.760><c> and</c><00:02:13.080><c> metadata</c><00:02:14.080><c> is</c><00:02:14.440><c> the</c><00:02:14.920><c> hardest</c>

00:02:15.300 --> 00:02:15.310 align:start position:0%
scaling and metadata is the hardest
 

00:02:15.310 --> 00:02:17.970 align:start position:0%
scaling and metadata is the hardest
problem<00:02:15.519><c> and</c><00:02:15.940><c> that</c><00:02:16.569><c> is</c><00:02:16.780><c> what</c><00:02:17.350><c> we</c><00:02:17.620><c> are</c><00:02:17.709><c> looking</c>

00:02:17.970 --> 00:02:17.980 align:start position:0%
problem and that is what we are looking
 

00:02:17.980 --> 00:02:20.990 align:start position:0%
problem and that is what we are looking
for<00:02:18.040><c> and</c><00:02:18.489><c> be</c><00:02:19.090><c> able</c><00:02:19.239><c> to</c><00:02:19.540><c> provide</c><00:02:20.080><c> an</c><00:02:20.440><c> actual</c>

00:02:20.990 --> 00:02:21.000 align:start position:0%
for and be able to provide an actual
 

00:02:21.000 --> 00:02:28.280 align:start position:0%
for and be able to provide an actual
scale<00:02:22.080><c> messaging</c><00:02:23.080><c> system</c><00:02:23.440><c> for</c><00:02:23.680><c> in</c><00:02:24.340><c> the</c><00:02:24.459><c> market</c>

00:02:28.280 --> 00:02:28.290 align:start position:0%
 
 

00:02:28.290 --> 00:02:30.920 align:start position:0%
 
[Music]

00:02:30.920 --> 00:02:30.930 align:start position:0%
[Music]
 

00:02:30.930 --> 00:02:33.780 align:start position:0%
[Music]
from<00:02:31.930><c> system</c><00:02:32.440><c> engineer</c><00:02:33.040><c> infrastructure</c>

00:02:33.780 --> 00:02:33.790 align:start position:0%
from system engineer infrastructure
 

00:02:33.790 --> 00:02:36.240 align:start position:0%
from system engineer infrastructure
engineer<00:02:34.239><c> perspective</c><00:02:34.530><c> we</c><00:02:35.530><c> write</c><00:02:35.799><c> a</c><00:02:36.010><c> lot</c><00:02:36.220><c> of</c>

00:02:36.240 --> 00:02:36.250 align:start position:0%
engineer perspective we write a lot of
 

00:02:36.250 --> 00:02:38.910 align:start position:0%
engineer perspective we write a lot of
code<00:02:36.510><c> around</c><00:02:37.510><c> I</c><00:02:37.810><c> think</c><00:02:38.260><c> a</c><00:02:38.440><c> synchronous</c>

00:02:38.910 --> 00:02:38.920 align:start position:0%
code around I think a synchronous
 

00:02:38.920 --> 00:02:41.880 align:start position:0%
code around I think a synchronous
processing<00:02:39.549><c> and</c><00:02:40.140><c> using</c><00:02:41.140><c> a</c><00:02:41.350><c> functional</c>

00:02:41.880 --> 00:02:41.890 align:start position:0%
processing and using a functional
 

00:02:41.890 --> 00:02:45.060 align:start position:0%
processing and using a functional
language<00:02:42.370><c> to</c><00:02:43.030><c> do</c><00:02:43.209><c> a</c><00:02:43.239><c> signal</c><00:02:44.230><c> processing</c><00:02:44.680><c> which</c>

00:02:45.060 --> 00:02:45.070 align:start position:0%
language to do a signal processing which
 

00:02:45.070 --> 00:02:47.550 align:start position:0%
language to do a signal processing which
simplified<00:02:46.060><c> the</c><00:02:46.329><c> implementation</c><00:02:47.049><c> design</c>

00:02:47.550 --> 00:02:47.560 align:start position:0%
simplified the implementation design
 

00:02:47.560 --> 00:02:50.340 align:start position:0%
simplified the implementation design
make<00:02:47.799><c> code</c><00:02:48.010><c> much</c><00:02:48.280><c> clearer</c><00:02:48.609><c> to</c><00:02:49.359><c> for</c><00:02:49.989><c> maintain</c>

00:02:50.340 --> 00:02:50.350 align:start position:0%
make code much clearer to for maintain
 

00:02:50.350 --> 00:02:53.599 align:start position:0%
make code much clearer to for maintain
and<00:02:50.590><c> that</c><00:02:51.190><c> is</c><00:02:51.370><c> the</c><00:02:51.730><c> most</c><00:02:52.660><c> beneficial</c>

00:02:53.599 --> 00:02:53.609 align:start position:0%
and that is the most beneficial
 

00:02:53.609 --> 00:02:55.470 align:start position:0%
and that is the most beneficial
advantage<00:02:54.609><c> that</c><00:02:54.940><c> we</c><00:02:55.030><c> can</c><00:02:55.180><c> get</c><00:02:55.359><c> from</c>

00:02:55.470 --> 00:02:55.480 align:start position:0%
advantage that we can get from
 

00:02:55.480 --> 00:03:03.750 align:start position:0%
advantage that we can get from
functional<00:02:56.019><c> program</c><00:02:56.620><c> for</c><00:02:56.859><c> for</c><00:02:57.190><c> systems</c>

00:03:03.750 --> 00:03:03.760 align:start position:0%
 
 

00:03:03.760 --> 00:03:07.539 align:start position:0%
 
I<00:03:04.760><c> think</c><00:03:05.480><c> it's</c><00:03:05.689><c> well</c><00:03:05.959><c> organized</c><00:03:06.620><c> the</c>

00:03:07.539 --> 00:03:07.549 align:start position:0%
I think it's well organized the
 

00:03:07.549 --> 00:03:10.289 align:start position:0%
I think it's well organized the
extensive<00:03:08.030><c> of</c><00:03:08.299><c> the</c><00:03:08.750><c> impressive</c><00:03:09.439><c> talks</c><00:03:09.829><c> and</c>

00:03:10.289 --> 00:03:10.299 align:start position:0%
extensive of the impressive talks and
 

00:03:10.299 --> 00:03:13.089 align:start position:0%
extensive of the impressive talks and
besides<00:03:11.299><c> I</c><00:03:11.719><c> work</c><00:03:11.930><c> for</c><00:03:12.200><c> Twitter</c><00:03:12.409><c> before</c><00:03:12.620><c> it's</c>

00:03:13.089 --> 00:03:13.099 align:start position:0%
besides I work for Twitter before it's
 

00:03:13.099 --> 00:03:21.620 align:start position:0%
besides I work for Twitter before it's
good<00:03:13.310><c> it's</c><00:03:13.909><c> good</c><00:03:14.090><c> for</c><00:03:14.269><c> coming</c><00:03:14.540><c> back</c>

00:03:21.620 --> 00:03:21.630 align:start position:0%
 
 

00:03:21.630 --> 00:03:26.409 align:start position:0%
 
[Music]