New amazon silk -cloud accelerated Web Browser

why we need new browser: time changed from earlier web to now web,devices, large content.
– mobile vs desktop (gap on loading page).
– tablet cannot process heavy duty graphics application unlike any desktop.
– tablet on cloud having EC2 instance (65 GB RAM,8core,optical network).
– split (on device, on cloud)
——————
decoupled element: dynamic split browsing:
backend cloud and front end browser:(backend does all optimisation)
optimisation at level of
networking (more processing on device less on cloud),
HTML,CSS,
collections(more processing on device less on cloud)
javascript
marshalling
native OM
formatting
block building
layout
display
—————-
When you click page from mobile device and click on another page on same site. what happens at backend.
1. Dns resolution–>find origin server.
2.TCP handshake
3.issue a request to server for web pages and related images and javascript.
4.(ask content u want) response comes back
5.acknowledgement for
cycle is repeated for each request.for every request cycle is repeated everytime no with split browser. devices uses wireless network.

In new backend Silk browser running on cloud.
(so many hops for request which takes about 100 millisecs per request compared with 10 millisecs of cloud internal response time)
-Persitent connection.
if all assets are on cloud (5 millisecs for each request).since the assets or pages are also living under same cloud. suppose request requires 80 files for a web page the difference adds up. to delay when user click and waiting for page to download.
———–
new silk browser indexes Page on cloud.
indexes the commonly used pages by you.
-with amazon ec2 cloud created a (limitless cache(store common files images,javascript,css) everyday)index on cloud for user(no storage on local storage)
– all storage on cloud.
-optimized content delivery.(so everything sits then client should not have situation like on cloud 50MB jpeg should not look like 3m jpeg on client)
——————————————————————————————————————————–
Machine learning :
detecting aggreate user behaviour pattern.(on cloud)
(predict user behaviour).
——————————————————————————————————————————–
– compution at cloud level
——————————————————————————————————————————–
New imporvements
1.-optimized last mile connection.(less time to hop on cloud then on web).
2-persistent connection.: seamless connection no delay in moving from one page to another
3-massive E2 server fleet.:EC2 instance on amazon cloud(65 GB RAM,8core,optical network)
4-page indexes.:indexing your behaviour daily on net.
5-advanced caching.: predictive proactive caching of data and pages
6-SSL security.-
7-image compression -.so image quality is maintained at client since everything is done at cloud only final output goes to client.
8-predictive rendering.- predictive analysis of user interest
9-machine learning.-finding patterns in user browsing.
10-encrypted delivery- secure transmission to ward off man in middle attack like senario.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s