This was extracted (@ 2018-06-20 20:10) from a list of minutes
which have been approved by the Board.
Please Note
The Board typically approves the minutes of the previous meeting at the
beginning of every Board meeting; therefore, the list below does not
normally contain details from the minutes of the most recent Board meeting.
2018 | 2017 | 2016 | 2015 | 2014 | 2013 | 2012 | 2011 | 2010 | 2009 | 2008 | 2007 | 2006 | 2005 | 2004 | 2003 | 2002 | 2001 | 2000 | 1999 | Pre-organization meetings
Report was filed, but display is awaiting the approval of the Board minutes.
Description:
Apache Mnemonic is an open-source Java library for durable object-oriented
programming on hybrid storage-class memory(e.g. NVM) space. it comes up
with durable object model (DOM) and durable computing model(DCM) and
takes full advantages of storage-class memory to simplify the code
complexity, avoid SerDe/(Un)Marshal, mitigate caching for constructing
next generation computing platform. Mnemonic makes the storing and
transmitting of massive linked objects graphs simpler and more efficient.
The performance tuning could also be mostly converged to a single point
of tuning place if based on Mnemonic to process and analyze
linked objects. The programmer is able to focus on durable object
oriented business logic instead of worrying about how to normalize/join,
serDe(un)marshal, cache and storing their linked business objects
with arbitrary complexity.
Issues:
There are no board-level issues at the moment.
Activity:
In this period of reporting, A new major improvement has been added to
Mnemonic codebase that helps the developer to remove the boilerplates
that are required to use durable classes with generic types.
Usually a custom factory proxy must be defined for each of durable
type if that is a generic type, right now, just need to instantiate a
helper class with 1 line code instead of 27 lines code;
this improvement makes code less error-prone and more concise.
With the support of some members of PMC, we may have a chance to run
Mnemonic on cutting edge server platform equipping with very large
volumn of non-volatile memory device. I think this may help us to
study the system and identify the gaps for durable models if any.
We also have fixed a major bug that may impact the user experience
on Fedora 27.
With the support of our community contributors, there are 2 conference
that has accepted our Mnemonic related topics, one is NVMW2018 as poster
and another is OSTS as presentation.
The Mnemonic code has become stable again and is ready for next release,
so the v0.11 might be released before next report.
Health Report:
Basically unchanged since the last report. Users are generally quiet
in public and no new users are appearing, but development continues.
PMC Changes:
- Currently 11 PMC members.
- No new PMC members added in the last 1 months.
Committer Base Changes:
- Currently 13 committers.
- No new committer added since the last committer on Jan. 8, 2018.
Releases:
- Last release was v0.10.0 on Mon Nov. 2017
- Still active development on next major version (0.11.0)
JIRA Activity:
- 13 JIRA tickets created since the last report (Feb. 2018)
- Also 13 JIRA tickets closed/resolved this period
Sincerely,
Gang(Gary) Wang on behalf of the Apache Mnemonic PMC
Description:
Apache Mnemonic is an open-source Java library for durable object-oriented
programming on hybrid storage-class memory(e.g. NVM) space. it comes up
with durable object model (DOM) and durable computing model(DCM) and
takes full advantages of storage-class memory to simplify the code
complexity, avoid SerDe/(Un)Marshal, mitigate caching for constructing
next generation computing platform. Mnemonic makes the storing and
transmitting of massive linked objects graphs simpler and more efficient.
The performance tuning could also be mostly converged to a single point
of tuning place if based on Mnemonic to process and analyze
linked objects. The programmer is able to focus on durable object
oriented business logic instead of worrying about how to normalize/join,
serDe(un)marshal, cache and storing their linked business objects
with arbitrary complexity.
Issues:
There are no board-level issues at the moment.
Activity:
In this period of reporting, we have introduced our project to Alibaba
JVM Team, one of our contributor presented the details of Mnemonic and
collected requirements of their business associated with our project.
We also have talked with China Telecom research team, they also got
interested in our project and trying to make use of Mnemonic for their
clustered computing platform, their leader also has provided some
feedback/suggestion and contributed some code to our project.
We feel that more and more industrial customers are beginning to learn
about how and what they can do for their system if non-volatile memory
become a generic device for their computing platforms soon.
Right now, the Durable Object/Computing models are quite new concepts
but we can expect that will be aware of and adopted quickly in near
future.
With the suggestion of one of our contributor, our community has worked
together to complete a document of Mnemonic tutorial, we also coded some
examples for the user to get started to learn Mnemonic.
We have voted a new committer to join our community, he is working for
constructing our website.
Health Report:
Basically unchanged since the last report. Users are generally quiet
in public and no new users are appearing, but development continues.
PMC Changes:
- Currently 11 PMC members.
- No new PMC members added in the last 1 months.
Committer Base Changes:
- Currently 13 committers.
- 1 new committer added on Jan. 8, 2018.
Releases:
- Last release was v0.10.0 on Mon Nov. 2017
- Still active development on next major version (0.11.0)
JIRA Activity:
- 15 JIRA tickets created since the last report (Jan. 2018)
- Also 10 JIRA tickets closed/resolved this period
Sincerely,
Gang(Gary) Wang on behalf of the Apache Mnemonic PMC
@Brett: discuss with PMC content of the report
Description:
Apache Mnemonic is an open-source Java library for durable object-oriented
programming on hybrid storage-class memory(e.g. NVM) space. it comes up
with durable object model (DOM) and durable computing model(DCM) and
takes full advantages of storage-class memory to simplify the code
complexity, avoid SerDe/(Un)Marshal, mitigate caching for constructing
next generation computing platform. Mnemonic makes the storing and
transmitting of massive linked objects graphs simpler and more efficient.
The performance tuning could also be mostly converged to a single point
of tuning place if based on Mnemonic to process and analyze
linked objects. The programmer are able to focus on durable object
oriented business logics instead of worrying about how to normalize/join,
serDe(un)marshal, cache and storing their linked business objects
with arbitrary complexity.
Issues:
There are no board-level issues at the moment.
Activity:
This is a vacation month so our community has no much activities durable
this month, basically, we have start to document tutorials, this task is
created by one of our contributor, and another one is to start making
Mnemonic supporting JDK9, the JDK9 changes something that our project
relies on, we have to migrate to JDK9 as well as be compatible with older
versions of JDK e.g. JDK8,JDK7. Lastly we have upvoted a new committer.
Health Report:
Basically unchanged since the last report. Users are generally quiet
in public and no new users are appearing, but development continues.
PMC Changes:
- Currently 11 PMC members.
- No new PMC members added in the last 1 months.
Committer Base Changes:
- Currently 12 committers.
- A new committer has elected and accepted to our committer base on Jan.08
Releases:
- Last release was v0.10.0 on Mon Nov. 2017
- Still active development on next major version (0.11.0)
JIRA Activity:
- 15 JIRA tickets created since the last report
- Also 3 JIRA tickets closed/resolved this period
Sincerely,
Gang(Gary) Wang on behalf of the Apache Mnemonic PMC
Description:
Apache Mnemonic is an open-source Java library for durable object-oriented
programming on hybrid storage-class memory(e.g. NVM) space. it comes up
with durable object model (DOM) and durable computing model(DCM) and
takes full advantages of storage-class memory to simplify the code
complexity, avoid SerDe/(Un)Marshal, mitigate caching for constructing
next generation computing platform. Mnemonic makes the storing and
transmitting of massive linked objects graphs simpler and more efficient.
The performance tuning could also be mostly converged to a single point
of tuning place if based on Mnemonic to process and analyze
linked objects. The programmer are able to focus on durable object
oriented business logics instead of worrying about how to normalize/join,
serDe(un)marshal, cache and storing their linked business objects
with arbitrary complexity.
Issues:
There are no board-level issues at the moment.
Activity:
Our project are just graduated from incubator so somethings are new to
us. we have a release since last report and one of our PMC found that
there are some issues caused by Java 9 compiling. we are going to get
insight into this problem and resolve it as soon as possible after fully
converted from incubating status. the major works that have been done
since last month are related to updating incubator associated
configuration, code and webpages. lastly, the adoption of storage-class
memory is not very straightforward and easy to integrated with existing
application and frameworks so we are trying to organize some other
Apache projects to form a joint working group for driving this technology
are able to be integrated smoothly.
The following people represented their project have got interested to
join this workgroup/program.
Ignite represented by Denis Magda
Arrow represented by Wes McKinney
Hbase represented by Anoop John
Crail represented by Patrick Stuedi
Hive/LLAP represented by Gopal Vijayaraghavan
ORC represented by Owen O'Malley
RocketMQ represented by Von Gosling
ActiveMQ represented by Clebert Suconic
Geode represented by Swapnil Bawaskar, Nilkanth Patel
the advanced storage-class memory technology will fundamentally change
the way to construct high performance applications similar to what
happened when replacing tape with disk technology since the 1980s. so we
advocate establishing an Apache working group to enhance the
collaboration and synergies for storage-class memory technology-oriented
projects.
To support this workgroup/program, we request to have a dedicate
mailing list, better if possible to be integrated with Slack.
Health Report:
Basically unchanged since the last report. Users are generally quiet
in public and no new users are appearing, but development continues.
PMC Changes:
- Currently 11 PMC members.
- No new PMC members added in the last 1 months.
Committer Base Changes:
- Currently 12 committers.
- No new changes to the committer base since last report.
Releases:
- Last release was v0.10.0 on Mon Nov. 2017
- Still active development on next major version (0.11.0)
JIRA Activity:
- 63 JIRA tickets created since last report (Sept. 2017)
- Also 47 JIRA tickets closed/resolved this period
Sincerely,
Gang(Gary) Wang on behalf of the Apache Mnemonic PMC
WHEREAS, the Board of Directors deems it to be in the best interests of the Foundation and consistent with the Foundation's purpose to establish a Project Management Committee charged with the creation and maintenance of open-source software, for distribution at no charge to the public, related to a transparent nonvolatile hybrid memory oriented library for Big data, High-performance computing, and Analytics. NOW, THEREFORE, BE IT RESOLVED, that a Project Management Committee (PMC), to be known as the "Apache Mnemonic Project", be and hereby is established pursuant to Bylaws of the Foundation, and be it further RESOLVED, that the Apache Mnemonic Project be and hereby is responsible for the creation and maintenance of software related to a transparent nonvolatile hybrid memory oriented library for Big data, High-performance computing and Analytics; and be it further RESOLVED, that the office of "Vice President, Apache Mnemonic" be and hereby is created, the person holding such office to serve at the direction of the Board of Directors as the chair of the Apache Mnemonic Project, and to have primary responsibility for management of the projects within the scope of responsibility of the Apache Mnemonic Project, and be it further RESOLVED, that the persons listed immediately below be and hereby are appointed to serve as the initial members of the Apache Mnemonic Project: * Andrew Kyle Purtell <apurtell@apache.org> * Debojyoti Dutta <ddutta@apache.org> * Gang Wang <garyw@apache.org> * Hao Cheng <chhao01@apache.org> * James R. Taylor <jamestaylor@apache.org> * Johnu George <johnu@apache.org> * Kai Zheng <drankye@apache.org> * Patrick D. Hunt <phunt@apache.org> * Rakesh Radhakrishnan <rakeshr@apache.org> * Uma Maheswara Rao G <umamahesh@apache.org> * Yanping Wang <yanpingw@apache.org> NOW, THEREFORE, BE IT FURTHER RESOLVED, that Gang Wang be appointed to the office of Vice President, Apache Mnemonic, to serve in accordance with and subject to the direction of the Board of Directors and the Bylaws of the Foundation until death, resignation, retirement, removal or disqualification, or until a successor is appointed, and be it further RESOLVED, that the initial Apache Mnemonic PMC be and hereby is tasked with the creation of a set of bylaws intended to encourage open development and increased participation in the Apache Mnemonic Project; and be it further RESOLVED, that the Apache Mnemonic Project be and hereby is tasked with the migration and rationalization of the Apache Incubator Mnemonic podling; and be it further RESOLVED, that all responsibilities pertaining to the Apache Incubator Mnemonic podling encumbered upon the Apache Incubator PMC are hereafter discharged. Special Order 7C, Establish the Apache Mnemonic Project, was approved by Unanimous Vote of the directors present.
Apache Mnemonic is a transparent NVM hybrid memory oriented library that has
created Durable Object Model (DOM), Durable Computing Model (DCM) and Durable Query Model (DQM)
that bring many advantages to significantly improve the performance of
massive real-time data processing/analytics. developers are able to use this
library to design their cache-less, SerDe-less and direct-native
high-performance applications or frameworks.
Mnemonic has been incubating since 2016-03-03.
Three most important issues to address in the move towards graduation:
1. Have elected new PPMC, Release Manager and committers
2. Have proposed Durable Query Model (DQM) based on DOM and DCM
3. Have prepared Apache maturity assessment checklist for review
Any issues that the Incubator PMC (IPMC) or ASF Board wish/need to be
aware of?
Not yet
How has the community developed since the last report?
Our last report was in Jun. Since then
* fulfilled the gaps from graduation
* elected new release manager
* elected new committers
* introduce owner of CQEngine to our community
* some contributors have started to vote for graduation
How has the project developed since the last report?
* improved Apache mnemonic official website
* implemented durable query in memory service level
* implemented pure Java volatile memory service
* improved the documentation of project specific release process
* released mnemonic-0.8.0-incubating
* started to vote mnemonic-0.9.0-incubating
How would you assess the podling's maturity?
Please feel free to add your own commentary.
[ ] Initial setup
[ ] Working towards first release
[ ] Community building
[X] Nearing graduation
[ ] Other:
Date of last release:
2017-06-23
When were the last committers or PPMC members elected?
Johnu George (PPMC on Aug. 04 2017)
PeiLi Shen (Committer on Aug. 13 2017)
Signed-off-by:
[X](mnemonic) Patrick Hunt
Comments: afaict ready to graduate (there is an active DISCUSS thread going on now)
[ ](mnemonic) Andrew Purtell
Comments:
[X](mnemonic) James Taylor
Comments:
[ ](mnemonic) Henry Saputra
Comments:
Apache Mnemonic is an advanced hybrid memory storage oriented library,
we've proposed a non-volatile/durable Java object model and durable
computing service that bring many advantages to significantly improve
the performance of massive real-time data
processing/analytics. developers are able to use this library to design
their cache-less, SerDe-less and native-direct computing high
performance applications.
Three most important issues to address in the move towards graduation:
1. Apache releases: we have release version 0.4.0-incubating
2. Integrated with Hadoop as Mapreduce Input/Output formatters.
3. Introduced Mnemonic to potential developers and users.
Any issues that the Incubator PMC (IPMC) or ASF Board wish/need to be
aware of?
No.
How has the community developed since the last report?
Our last report was in Dec. Since then
* Two more developers are going to contribute code.
* We need to encourage more discussion on the dev list in the coming months
How has the project developed since the last report?
* Upgraded MemChunkHolder/MemBufferHolder to DurableChunk/DurableHolder
* Provided mnemonic-hadoop-mapreduce module with testcases
* and other bugfixes, features, improvements
How would you assess the podling's maturity?
Please feel free to add your own commentary.
[ ] Initial setup
[ ] Working towards first release
[ ] Community building
[*] Nearing graduation
[ ] Other:
Date of last release:
2017-01-24
When were the last committers or PMC members elected?
None
Signed-off-by:
[X](mnemonic) Patrick Hunt
Comments:
One year milestone as podling this month. Community growth has been
slow and activity dominated by one to two committers. Not clear how
active the mentors are in helping the podling - only two mentors
signed off on the report last quarter.
[X](mnemonic) Andrew Purtell
Comments:
Mentors have not been active, including myself. We may want to
solicit new or additional mentors for this podling.
[ ](mnemonic) James Taylor
Comments:
[ ](mnemonic) Henry Saputra
Comments:
Apache Mnemonic is an advanced hybrid memory storage oriented library,
we've proposed a non-volatile/durable Java object model and durable
computing service that bring several advantages to significantly improve
the performance of massive real-time data
processing/analytics. developers are able to use this library to design
their cache-less, SerDe-less and native-direct computing high
performance applications.
Three most important issues to address in the move towards graduation:
1. Apache releases: we have release version 0.3.0-incubating
2. Improved dev. docker to facilitate development.
3. provided information to potential customers e.g. bank user, other
project's Apache committers
Any issues that the Incubator PMC (IPMC) or ASF Board wish/need to be
aware of?
No.
How has the community developed since the last report?
Our last report was in April. Since then
* Two more developers are going to join our community.
* We need to encourage more discussion on the dev list in the coming months
How has the project developed since the last report?
* Added durable native sorting service (n-dim linked data) with
testcases
* Removed the 3rd Maven repository and use Maven central repository
only
* DOM and DNCS could apply to volatile memory service except fixed k-v
store
* Added a bench workload to verify the benefits from DNCS with
scripts, generator
* Added a sys-vmem service that makes it possible to utilize system
memory for DNCS in an unified way
* Added a script to bundle signed portable artifacts for maven
repository submitting
* and other bugfixes, features, improvements
Date of last release:
2016-10-21
When were the last committers or PMC members elected?
None
Signed-off-by:
[X](mnemonic) Patrick Hunt
[ ](mnemonic) Andrew Purtell
[ ](mnemonic) James Taylor
[X](mnemonic) Henry Saputra
Apache Mnemonic is an advanced hybrid memory storage oriented library, we've
proposed a non-volatile/durable Java object model and durable computing
service that bring several advantages to significantly improve the
performance of massive real-time data processing/analytics. developers are
able to use this library to design their cache-less and SerDe-less high
performance applications.
Three most important issues to address in the move towards graduation:
1. Apache releases: we have release version 0.2.0-incubating
2. Encouraging community participation and contribution
3. Invented Durable Computing Model and Service on top of Durable Object
Model
Any issues that the Incubator PMC (IPMC) or ASF Board wish/need to be
aware of?
No.
How has the community developed since the last report?
Our last report was in April. Since then
* Twelve new contributors have joined and seven contributors have submitted
their patches.
* We need to encourage more discussion on the dev list in the coming months
How has the project developed since the last report?
* Added memory service and testcases
* Added computing service and testcases
* Re-organized sub-modules
* Added Apache Rat and Checkstyle to ensure compliance
* Added release.sh, runall.sh script tools
* Supported Docker dev. construction
* Removed the dependency of customized 3rd party library
* Removed the dependency of special version of nvml
* Constructed project website (WIP)
* Supported discrete addressing for memory service
* and other bugfixes, features, improvements
Date of last release:
2016-08-02
When were the last committers or PMC members elected?
None
Signed-off-by:
[X](mnemonic) Patrick Hunt
[X](mnemonic) Andrew Purtell
[x](mnemonic) James Taylor
[ ](mnemonic) Henry Saputra
Mnemonic is a Java based non-volatile memory library for in-place structured data processing and computing. Mnemonic has been incubating since 2016-03-03. On 05/24/2016, we released our first official build at https://dist.apache.org/repos/dist/release/incubator/mnemonic/0.1.2-incubating/ Various metrics for May are as follows: (5/3-5/31/2016) +------------------------------------------+ | Metric | counts | +------------------------------------------+ | Non Merge Commits | 23 | | Contributors | 3 | | Jira New Issues | 18 | | Resolved Issues | 18 | | Pull Requests merged | 0 | | Pull Requests proposed | 0 | +------------------------------------------+ Three most important issues to address in the move towards graduation: 1. Apache releases: Community voted and released Mnemonic-0.1.2-incubating 2. Expand and build strong community, keep active dev list discussions 3. Develop new features: We are working in Native computing layer design Any issues that the Incubator PMC (IPMC) or ASF Board wish/need to be aware of? There are no issues requiring board attention at this time How has the community developed since the last report? Yanping and Gary presented Mnemonic project at ApacheCon Big Data Conference. Yanping presented at In-Memory Computing Summit. Both talks were well received. Several developers approached us to ask additional details on how Mnemonic worked for Spark. They are interested to see if we are able to successfully integrate Mnemonic into other projects. While we are focusing on develop new features, we will try to contact other projects' communities to explore integration opportunities. As the official release is done this month, for the next 3 months we expect integration activities will be increased. How has the project developed since the last report? We have been focusing the first Apache release over the past weeks. We also explored ideas on design native computing layer which will further enhance durable Java object model on large non-volatile memory. Date of last release: 2016-05-24 When were the last committers or PMC members elected? No election, the project is still in early stage of developing. Signed-off-by: [X](mnemonic) Patrick Hunt [X](mnemonic) Andrew Purtell [ ](mnemonic) James Taylor [ ](mnemonic) Henry Saputra
Mnemonic is a Java based non-volatile memory library for in-place structured
data processing and computing.
Mnemonic has been incubating since 2016-03-03. Mnemonic code was moved into
ASF on 3/24.
Various metrics for April are as follows:
+--------------------------------------------+
| Metric | counts |
+--------------------------------------------+
| Non Merge Commits | 16 |
| Contributors | 4 |
| Jira New Issues | 20 |
| Resolved Issues | 24 |
| Pull Requests merged | 6 |
| Pull Requests proposed | 7 |
+--------------------------------------------+
Three most important issues to address in the move towards graduation:
1. Apache releases: we are preparing 0.1.0 release
2. Expand and build strong community, keep active dev list discussions,
adding new contributors
a. Mnemonic website is up: http://mnemonic.incubator.apache.org/
b. Mnemonic cwiki is up:
https://cwiki.apache.org/confluence/display/MNEMONIC/Apache+Mnemonic+Ho
me
3. Develop new features: We are working in Native computing layer design.
Any issues that the Incubator PMC (IPMC) or ASF Board wish/need to be aware
of?
Everything is on track, nothing urgent at this time.
How has the community developed since the last report?
1. We completed setting up infrastructure. Both project website and cwiki
are up running. We are working on adding contents and documentation to
the sites so new contributors and community members can get better view
of what this project is and how to get involved.
2. We realized, unlike other application level projects, library (or API)
projects traditionally had less contributors until certain amount of
adopters successfully integrated those libraries. We plan to help some
integrations, such as HBase, Arrow, etc. after our first official code
release.
3. We submitted 2 proposals for Strata Hadoop World NYC.
4. We will present Mnemonic project in upcoming ApacheCon Big Data
conference on 5/11.
How has the project developed since the last report?
We are designing the native computing infrastructure, built native
computing interface code.
In preparing first official release:
1. Defined coding guidelines for the project. Reformatted all code based
on defined coding style.
2. Added the checkstyle to automate the checking process.
3. Formalized the names of annotations, classes and methods.
4. Removed the dependency of property mnemonic.core.version.
5. Added a new profile for artifacts releasing and signing.
6. Added release plugin and configuration for releasing.
Date of last release:
No release yet, working on 0.1.0 release
When were the last committers or PMC members elected?
No election, since the project is still in early stage of developing.
Signed-off-by:
[X](mnemonic) Patrick Hunt
[X](mnemonic) Andrew Purtell
[x](mnemonic) James Taylor
[ ](mnemonic) Henry Saputra
Mnemonic is a Java based non-volatile memory library for in-place structured
data processing and computing.
Mnemonic has been incubating since 2016-03-03.
Mnemonic code was moved into ASF on 3/24
Various metrics for March are as follows:
+--------------------------------------------+
| Metric | counts |
+--------------------------------------------+
| Non Merge Commits | 82 |
| Contributors | 10 |
| Jira New Issues | 13 |
| Resolved Issues | 4 |
| Pull Requests merged | 8 |
| Pull Requests proposed | 8 |
+--------------------------------------------+
Three most important issues to address in the move towards graduation:
1. Apache releases: we are preparing 0.1.0 release
2. Expand and build strong community, keep active dev list discussions,
adding new contributors
3. Develop new features
Any issues that the Incubator PMC (IPMC) or ASF Board wish/need to be
aware of?
Everything is on track, nothing urgent at this time.
How has the community developed since the last report?
The project is in early stage in setting up the infra. In addition to
contributors listed in Mnemonic proposal, we had one new contributor
joined us. We are actively discuss with Arrow community in dev list for
the idea on integration. Apache Arrow community is very interested in
Mnemonic integration for extending Arrow into persistent memory and
storage.
How has the project developed since the last report?
We ported source code in Apache on 3/24 after SGA was filed on 3/23. We
are doing several refactorings and defining coding guidelines to fit
Apache project standard.
We are working on enabling Apache Arrow to use Mnemonic as underlying
pooling mechanism.
We are designing the native computing infrastructure, built native
computing interface code.
Bug tracking JIRA setup was done, 13 issues were filed. 4 issues were
solved.
Mailing list discussions are happening actively related to infrastructure,
code standard, features for the project.
Date of last release:
No release yet, working on 0.1.0 release
When were the last committers or PMC members elected?
No election, since the project is just starting. As part of incubation
process, initial committers and contributors were proposed and added.
Signed-off-by:
[X](mnemonic) Patrick Hunt
[X](mnemonic) Andrew Purtell
[X](mnemonic) James Taylor
[X](mnemonic) Henry Saputra