Author: jamestaylor
Date: Tue Mar 21 05:43:24 2017
New Revision: 1787899
URL: http://svn.apache.org/viewvc?rev=1787899&view=rev
Log:
Update formatting for Sogou
Modified:
phoenix/site/publish/language/datatypes.html
phoenix/site/publish/language/functions.html
phoenix/site/publish/language/index.html
phoenix/site/publish/who_is_using.html
phoenix/site/source/src/site/markdown/who_is_using.md
Modified: phoenix/site/publish/language/datatypes.html
URL:
http://svn.apache.org/viewvc/phoenix/site/publish/language/datatypes.html?rev=1787899&r1=1787898&r2=1787899&view=diff
==============================================================================
--- phoenix/site/publish/language/datatypes.html (original)
+++ phoenix/site/publish/language/datatypes.html Tue Mar 21 05:43:24 2017
@@ -1,7 +1,7 @@
<!DOCTYPE html>
<!--
- Generated by Apache Maven Doxia at 2017-03-14
+ Generated by Apache Maven Doxia at 2017-03-20
Rendered using Reflow Maven Skin 1.1.0
(http://andriusvelykis.github.io/reflow-maven-skin)
-->
<html xml:lang="en" lang="en">
Modified: phoenix/site/publish/language/functions.html
URL:
http://svn.apache.org/viewvc/phoenix/site/publish/language/functions.html?rev=1787899&r1=1787898&r2=1787899&view=diff
==============================================================================
--- phoenix/site/publish/language/functions.html (original)
+++ phoenix/site/publish/language/functions.html Tue Mar 21 05:43:24 2017
@@ -1,7 +1,7 @@
<!DOCTYPE html>
<!--
- Generated by Apache Maven Doxia at 2017-03-14
+ Generated by Apache Maven Doxia at 2017-03-20
Rendered using Reflow Maven Skin 1.1.0
(http://andriusvelykis.github.io/reflow-maven-skin)
-->
<html xml:lang="en" lang="en">
Modified: phoenix/site/publish/language/index.html
URL:
http://svn.apache.org/viewvc/phoenix/site/publish/language/index.html?rev=1787899&r1=1787898&r2=1787899&view=diff
==============================================================================
--- phoenix/site/publish/language/index.html (original)
+++ phoenix/site/publish/language/index.html Tue Mar 21 05:43:24 2017
@@ -1,7 +1,7 @@
<!DOCTYPE html>
<!--
- Generated by Apache Maven Doxia at 2017-03-14
+ Generated by Apache Maven Doxia at 2017-03-20
Rendered using Reflow Maven Skin 1.1.0
(http://andriusvelykis.github.io/reflow-maven-skin)
-->
<html xml:lang="en" lang="en">
Modified: phoenix/site/publish/who_is_using.html
URL:
http://svn.apache.org/viewvc/phoenix/site/publish/who_is_using.html?rev=1787899&r1=1787898&r2=1787899&view=diff
==============================================================================
--- phoenix/site/publish/who_is_using.html (original)
+++ phoenix/site/publish/who_is_using.html Tue Mar 21 05:43:24 2017
@@ -1,7 +1,7 @@
<!DOCTYPE html>
<!--
- Generated by Apache Maven Doxia at 2017-03-14
+ Generated by Apache Maven Doxia at 2017-03-20
Rendered using Reflow Maven Skin 1.1.0
(http://andriusvelykis.github.io/reflow-maven-skin)
-->
<html xml:lang="en" lang="en">
@@ -164,15 +164,6 @@
<tbody>
<tr class="b"></tr>
<tr class="a">
- <td> <img src="images/using/bb.png" alt="" /> <br /><br /> At
Bloomberg, patterns of access to financial datasets are diverse and complex.
HBase provides the scalability and strong consistency that our use cases
demand. However, we need more than a key-value store. We need ANSI SQL to
reduce the barriers to adoption, we need features such as secondary indices to
support lookups along multiple axes and cursors to handle UI pagination. Apache
Phoenix provides a rich set of capabilities above HBase that makes it a
critical piece of our data platform. <br /><br /> Saurabh Agarwal, Bloomberg
Data Platform </td>
- </tr>
- <tr class="b">
- <td> <img src="images/using/eharmony.png" alt="" /> <br /><br /> At
eHarmony, Apache Phoenix serves as an SQL abstraction for the HBase storage
where we maintain details about potential relationship matches identified for
our users. We store presentation-ready user match feeds in HBase, and serve the
data to one of the most visited pages on <a class="externalLink"
href="http://www.eharmony.com";>eharmony.com</a>. Apache Phoenix helped us to
build a query abstraction layer that eased our development process, enabling us
to to apply various filters and sorting on the aggregated data in the HBase
store.
- <blockquote>
- "The integration with Apache Phoenix has not only stabilized our
system, but also reduced response time for loading hundreds of matches on a
page to below 200ms."
- </blockquote> <br /> Vijay Vangapandu, Principal Software Engineer
</td>
- </tr>
- <tr class="a">
<td> <img src="images/using/sf.png" alt="" /> <br /><br /> In our
Force.com platform, we rely on Apache Phoenix to run interactive queries
against big data residing in HBase leveraging <br /><br />
<ul>
<li> multi-tenant tables for customization and scale out across our
diverse customer schemas</li>
@@ -184,6 +175,15 @@
</blockquote> <br /> Steven Tamm, CTO </td>
</tr>
<tr class="b">
+ <td> <img src="images/using/bb.png" alt="" /> <br /><br /> At
Bloomberg, patterns of access to financial datasets are diverse and complex.
HBase provides the scalability and strong consistency that our use cases
demand. However, we need more than a key-value store. We need ANSI SQL to
reduce the barriers to adoption, we need features such as secondary indices to
support lookups along multiple axes and cursors to handle UI pagination. Apache
Phoenix provides a rich set of capabilities above HBase that makes it a
critical piece of our data platform. <br /><br /> Saurabh Agarwal, Bloomberg
Data Platform </td>
+ </tr>
+ <tr class="a">
+ <td> <img src="images/using/eharmony.png" alt="" /> <br /><br /> At
eHarmony, Apache Phoenix serves as an SQL abstraction for the HBase storage
where we maintain details about potential relationship matches identified for
our users. We store presentation-ready user match feeds in HBase, and serve the
data to one of the most visited pages on <a class="externalLink"
href="http://www.eharmony.com";>eharmony.com</a>. Apache Phoenix helped us to
build a query abstraction layer that eased our development process, enabling us
to to apply various filters and sorting on the aggregated data in the HBase
store.
+ <blockquote>
+ "The integration with Apache Phoenix has not only stabilized our
system, but also reduced response time for loading hundreds of matches on a
page to below 200ms."
+ </blockquote> <br /> Vijay Vangapandu, Principal Software Engineer
</td>
+ </tr>
+ <tr class="b">
<td> <img src="images/using/hw.png" alt="" /> <br /><br /> Hortonworks
supports Apache Phoenix as a feature rich ANSI SQL interface for Apache HBase
in Hortonworks Data Platform (HDP). It plays a critical role for our customers
who want diverse choice for data access in Hadoop and want a simple interface
to build low-latency, large scale applications. Critical features, such as
secondary indexing have made Phoenix the API of choice for building these HBase
applications. <br /><br /> Devaraj Das, Cofounder <br /> </td>
</tr>
<tr class="a">
@@ -207,6 +207,9 @@
<li> With deep analytics, the company can make more intelligent
decisions, and analyze performance and alignment with business objectives</li>
</ul> Sudhir Kulkarni, VP of Data and Analytics <br /> </td>
</tr>
+ <tr class="a">
+ <td> <img src="images/using/dp.png" alt="" /> <br /><br /> At Delta
Projects we use Phoenix for storing data as a basis for measuring activities
and generating reports. We chose Phoenix because it provides the scalability of
HBase and the expressiveness of SQL.<br /><br /> Kristoffer Sjögren,
System Developer <br /></td>
+ </tr>
</tbody>
</table>
<!-- End First Column --> </td>
@@ -219,43 +222,44 @@
<td></td>
</tr>
<tr class="a">
- <td><br /> <img src="images/using/homeaway.png" alt="" /> <br /><br />
Apache Phoenix enables easy integration with HBase for systems that rely on
JDBC/SQL. HomeAway, the world leader in Vacation Rentals, leverages Phoenix as
a SQL abstraction for HBase's powerful columnar storage to generate statistics
for vacation rental owners on HomeAway's Owner Dashboard. These statistics help
HomeAway vacation rental owners gain key insights about the performance of
their vacation rental, how well it is doing against 'the market', and how well
it is doing historically.<br /><br /> From a pool of billions of records that
go back 2 years, HomeAway is able to serve up customer-facing webpages from
HBase, using Phoenix, in less than a second for the majority of our vacation
rental owners. With Phoenix and HBase, HomeAway is able to share the same
insight it has internally on the vacation rental market to its owners
empowering them with the necessary data to make the right decisions maximizi
ng their return on their vacation rental investment.<br /><br /> René
X. Parra, Principal Architect <br /></td>
+ <td> <br /> <img src="images/using/sogou.png" alt="" /> <br /><br /> We
adopted Apache Phoenix since 2015, mainly for two scenarios: <br /> <br />
+ <ol style="list-style-type: decimal">
+ <li>Business Intelligence: We use HBase+Phoenix to store billion
records of our Ad Exchange, thanks to the SQL abstraction and secondary indexes
of Phoenix, we can provide multidimensional statistical and analytical reports
to our advertisers, empowering them with thorough insight to make the
intelligent decisions maximizing their investment revenue.</li>
+ <li>Technology Infrastructureï¼ Our Monitoring Platform and
Distributed Service Tracing Platform uses HBase+Phoenix to continuously collect
various metrics and logs(about 100k records per second at present) ,and with
the high performance of Phoenix we can easily generate statistics for our
system operation health measurement and service dependency analysis.</li>
+ </ol> Cheng Lei, Infrastructure Software Enginner<br /></td>
</tr>
<tr class="b">
- <td> <img src="images/using/ss.png" alt="" /> <br /><br /> At Sift
Science we use Phoenix to power our OLAP infrastructure. This influences our
machine learning feature engineering which is critical in the model training
pipeline. Having a simple SQL-based interface also allows us to expose data
insights outside of the engineering organization. Finally, running Phoenix on
top of our existing HBase infrastructure gives us the ability to scale our
ad-hoc query needs. <br /><br /> Andrey Gusev, Tech Lead, Machine Learning
Infrastructure </td>
+ <td><br /> <img src="images/using/homeaway.png" alt="" /> <br /><br />
Apache Phoenix enables easy integration with HBase for systems that rely on
JDBC/SQL. HomeAway, the world leader in Vacation Rentals, leverages Phoenix as
a SQL abstraction for HBase's powerful columnar storage to generate statistics
for vacation rental owners on HomeAway's Owner Dashboard. These statistics help
HomeAway vacation rental owners gain key insights about the performance of
their vacation rental, how well it is doing against 'the market', and how well
it is doing historically.<br /><br /> From a pool of billions of records that
go back 2 years, HomeAway is able to serve up customer-facing webpages from
HBase, using Phoenix, in less than a second for the majority of our vacation
rental owners. With Phoenix and HBase, HomeAway is able to share the same
insight it has internally on the vacation rental market to its owners
empowering them with the necessary data to make the right decisions maximizi
ng their return on their vacation rental investment.<br /><br /> René
X. Parra, Principal Architect <br /></td>
</tr>
<tr class="a">
+ <td> <img src="images/using/ss.png" alt="" /> <br /><br /> At Sift
Science we use Phoenix to power our OLAP infrastructure. This influences our
machine learning feature engineering which is critical in the model training
pipeline. Having a simple SQL-based interface also allows us to expose data
insights outside of the engineering organization. Finally, running Phoenix on
top of our existing HBase infrastructure gives us the ability to scale our
ad-hoc query needs. <br /><br /> Andrey Gusev, Tech Lead, Machine Learning
Infrastructure </td>
+ </tr>
+ <tr class="b">
<td> <img src="images/using/ab.png" alt="" /> <br /><br /> At Alibaba
there're two main scenarios of using Phoenix:<br /><br />
<ol style="list-style-type: decimal">
<li> Large dataset with relatively small result set, say 10 thousands
of records or so. We choose to use Phoenix in this kind of scenario because
it's much more easier for user to use than HBase native api, meantime it
supports orderby/groupby syntax</li>
<li> Large dataset with large result set, it might be millions of
records in the result set even after PrimaryKey filter, and often along with
lots of aggregation/orderby/groupby invocation. We choose to use Pheonix in
this kind of scenario because Pheonix makes it possible to do complicated query
in HBase, and it supports more and more features in traditional DB like oracle,
which makes it much more easier for our user to migrate there BI query onto
HBase</li>
</ol> Jaywong, Software Engineer <br /></td>
</tr>
- <tr class="b">
+ <tr class="a">
<td> <img src="images/using/ebay.png" alt="" /> <br /><br /> We have
been exploring Phoenix since July, 2014 and have successfully achieved couple
of analytics use cases with huge data set. We were able to achieve read/write
performance in ms even slicing and dicing data in many dimensions.<br /><br />
<ol style="list-style-type: decimal">
<li> Path or Flow analysis<br /> This use case was very specific and
targeted for core mobile native apps where we were trying to find user behavior
with many dimension App, Version, device , OS version, carrier etc. This was
offline process where we process and aggregate daily data and load once in
phoenix schema.</li>
<li> Real Time analytics data trend.<br /> This is near real time
aggregation of tracking data to find trend of events with multi-dimensional. It
does write aggregated data to hBase + Phoenix continuously (at present
12k-15k/s records) and read for report generation at the same time.</li>
</ol> Jogendar Singh, Engineering Manager, Mobile Platform <br /></td>
</tr>
- <tr class="a">
+ <tr class="b">
<td><br /> <img src="images/using/ng.png" alt="" /> <br /><br /> Apache
Phoenix allows users of our customer analytics platform Lily to easily ingest
and manage customer fact data. Our users don't have to learn complex or
specific APIs for this, but can tap into a familiar competence: SQL. NGDATA is
happy to both use and contribute to the Apache Phoenix project, which proves to
be a solid choice backed by a great community on a day-to-day basis.<br /><br
/> Steven Noels, CTO <br /></td>
</tr>
- <tr class="b">
+ <tr class="a">
<td> <img src="images/using/pn.png" alt="" /> <br /><br /> Apache
Phoenix has helped us load and query hundreds of billions of records. The
salting and secondary indexes have saved considerable development time and the
SQL interface has been an easy entry point for developers.<br /><br /> Ralph
Perko, Software Architect/Developer <br /></td>
</tr>
- <tr class="a">
+ <tr class="b">
<td> <img src="images/using/sb.png" alt="" /> <br /><br /> At
Socialbakers we use Phoenix for on demand data aggregations. Because of the
floating time range of our custom reports we aggregate hundreds of megabytes
per request on the server side. Phoenix can handle those requests with low
latency and high throughput. It also provides an easy to use SQL interface and
helps us build scalable and highly available applications quickly and
reliably.<br /><br />
<blockquote>
"For us, the most valuable feature is the out of the box server
side aggregations."
</blockquote><br /> Martin Homolka, CTO <br /></td>
</tr>
- <tr class="b">
- <td> <img src="images/using/dp.png" alt="" /> <br /><br /> At Delta
Projects we use Phoenix for storing data as a basis for measuring activities
and generating reports. We chose Phoenix because it provides the scalability of
HBase and the expressiveness of SQL.<br /><br /> Kristoffer Sjögren,
System Developer <br /></td>
- </tr>
- <tr class="a">
- <td> <br /> <img src="images/using/sogou.png" alt="" /> <br /><br /> We
adopted Apache Phoenix since 2015, mainly for two scenarios: 1.Business
Intelligence: We use HBase+Phoenix to store billion records of our Ad Exchange,
thanks to the SQL abstraction and secondary indexes of Phoenix, we can provide
multidimensional statistical and analytical reports to our advertisers,
empowering them with thorough insight to make the intelligent decisions
maximizing their investment revenue.<br /><br /> 2.Technology Infrastructureï¼
Our Monitoring Platform and Distributed Service Tracing Platform uses
HBase+Phoenix to continuously collect various metrics and logs(about 100k
records per second at present) ,and with the high performance of Phoenix we can
easily generate statistics for our system operation health measurement and
service dependency analysis.<br /><br /> Cheng Lei, Infrastructure Software
Enginner<br /></td>
- </tr>
</tbody>
</table>
<!-- End Second Column --> </td>
Modified: phoenix/site/source/src/site/markdown/who_is_using.md
URL:
http://svn.apache.org/viewvc/phoenix/site/source/src/site/markdown/who_is_using.md?rev=1787899&r1=1787898&r2=1787899&view=diff
==============================================================================
--- phoenix/site/source/src/site/markdown/who_is_using.md (original)
+++ phoenix/site/source/src/site/markdown/who_is_using.md Tue Mar 21 05:43:24
2017
@@ -10,6 +10,24 @@
<tr></tr>
<tr><td>
+<img src="images/using/sf.png"/>
+<br/><br/>
+In our Force.com platform, we rely on Apache Phoenix to run
+interactive queries against big data residing in HBase leveraging
+<br/><br/>
+<ul>
+<li> multi-tenant tables for customization and scale out across our
+diverse customer schemas</li>
+<li> aggregation to build roll-up summaries</li>
+<li> secondary indexes to improve performance</li>
+</ul>
+<blockquote>"Apache Phoenix is the foundation of our big data stack, allowing
us
+to run interactive queries against HBase data in a performant
manner."</blockquote>
+<br/>
+Steven Tamm, CTO
+</td></tr>
+
+<tr><td>
<img src="images/using/bb.png"/>
<br/><br/>
At Bloomberg, patterns of access to financial datasets are diverse and
complex. HBase provides the scalability and strong consistency that our use
cases demand. However, we need more than a key-value store. We need ANSI SQL to
reduce the barriers to adoption, we need features such as secondary indices to
support lookups along multiple axes and cursors to handle UI pagination. Apache
Phoenix provides a rich set of capabilities above HBase that makes it a
critical piece of our data platform.
@@ -34,24 +52,6 @@ Vijay Vangapandu, Principal Software Eng
</td></tr>
<tr><td>
-<img src="images/using/sf.png"/>
-<br/><br/>
-In our Force.com platform, we rely on Apache Phoenix to run
-interactive queries against big data residing in HBase leveraging
-<br/><br/>
-<ul>
-<li> multi-tenant tables for customization and scale out across our
-diverse customer schemas</li>
-<li> aggregation to build roll-up summaries</li>
-<li> secondary indexes to improve performance</li>
-</ul>
-<blockquote>"Apache Phoenix is the foundation of our big data stack, allowing
us
-to run interactive queries against HBase data in a performant
manner."</blockquote>
-<br/>
-Steven Tamm, CTO
-</td></tr>
-
-<tr><td>
<img src="images/using/hw.png"/>
<br/><br/>
Hortonworks supports Apache Phoenix as a feature rich ANSI SQL
@@ -141,6 +141,17 @@ Sudhir Kulkarni, VP of Data and Analytic
<br/>
</td></tr>
+<tr><td>
+<img src="images/using/dp.png"/>
+<br/><br/>
+
+At Delta Projects we use Phoenix for storing data as a basis for
+measuring activities and generating reports. We chose Phoenix because
+it provides the scalability of HBase and the expressiveness of SQL.<br/><br/>
+
+Kristoffer Sjögren, System Developer
+<br/></td></tr>
+
</table>
<!--End First Column-->
</td>
@@ -151,6 +162,21 @@ Sudhir Kulkarni, VP of Data and Analytic
<tr style="display: none;"><td></td></tr>
+<tr><td>
+<br/>
+<img src="images/using/sogou.png"/>
+<br/><br/>
+
+We adopted Apache Phoenix since 2015, mainly for two scenarios:
+<br/>
+<br/>
+<ol>
+<li>Business Intelligence: We use HBase+Phoenix to store billion records of
our Ad Exchange, thanks to the SQL abstraction and secondary indexes of
Phoenix, we can provide multidimensional statistical and analytical reports to
our advertisers, empowering them with thorough insight to make the intelligent
decisions maximizing their investment revenue.</li>
+<li>Technology Infrastructureï¼ Our Monitoring Platform and Distributed
Service Tracing Platform uses HBase+Phoenix to continuously collect various
metrics and logs(about 100k records per second at present) ,and with the high
performance of Phoenix we can easily generate statistics for our system
operation health measurement and service dependency analysis.</li>
+</ol>
+
+Cheng Lei, Infrastructure Software Enginner<br/></td></tr>
+
<tr><td><br/>
<img src="images/using/homeaway.png"/>
<br/><br/>
@@ -288,27 +314,6 @@ aggregations."</blockquote><br/>
Martin Homolka, CTO
<br/></td></tr>
-<tr><td>
-<img src="images/using/dp.png"/>
-<br/><br/>
-
-At Delta Projects we use Phoenix for storing data as a basis for
-measuring activities and generating reports. We chose Phoenix because
-it provides the scalability of HBase and the expressiveness of SQL.<br/><br/>
-
-Kristoffer Sjögren, System Developer
-<br/></td></tr>
-
-<tr><td>
-<br/>
-<img src="images/using/sogou.png"/>
-<br/><br/>
-
-We adopted Apache Phoenix since 2015, mainly for two scenarios:
-1.Business Intelligence: We use HBase+Phoenix to store billion records of our
Ad Exchange, thanks to the SQL abstraction and secondary indexes of Phoenix,
we can provide multidimensional statistical and analytical reports to our
advertisers, empowering them with thorough insight to make the intelligent
decisions maximizing their investment revenue.<br/><br/>
-2.Technology Infrastructureï¼ Our Monitoring Platform and Distributed
Service Tracing Platform uses HBase+Phoenix to continuously collect various
metrics and logs(about 100k records per second at present) ,and with the high
performance of Phoenix we can easily generate statistics for our system
operation health measurement and service dependency analysis.<br/><br/>
-
-Cheng Lei, Infrastructure Software Enginner<br/></td></tr>
</table>
<!--End Second Column-->
