Welcome!

Websphere Authors: Trevor Parsons, Pat Romanski, Carmen Gonzalez, Elizabeth White, Liz McMillan

Related Topics: Websphere

Websphere: Article

Managing JDBC Connections

Managing JDBC Connections

If application servers like WebSphere are the heart of e-business, then the database is surely the lifeblood. Whether your firm is managing thousands of airline flights a day or simply exposing a library's card catalog to the Web, you must deal with database connectivity.

Java developers have a simple way to manage database connectivity at their disposal: JDBC. It provides a convenient interface to access a variety of vendors' databases via JDBC drivers. Simply plug in a new driver and your application can talk to DB2, Oracle, Sybase, Informix, and others.

The JDBC version 2.0 extensions introduced a feature called connection pooling, providing better performance and scalability. IBM implemented a JDBC 2.0-compliant connection pool in WebSphere 3.5, and continues to enhance it in version 4.0.

In addition to connection pooling, the WebSphere connection manager offers many additional qualities to enhance an application's performance and scalability for the savvy Web developer. This article will highlight those features and show how they can be used effectively for building better Web applications. In addition, this article will describe many best practices and potential pitfalls for developing with WebSphere's connection manager.

While the following information will provide an improvement to applications, note that this is meant to aid in application design, not replace it. No amount of "tweaking" can make up for a poor design, but a well-designed application with poor implementation of database connectivity logic can be improved. This article will address the latter case.

Background
The WebSphere connection manager has many facets. At the core, however, it's simply a connection pool. Object pools are nothing new.

The basic reason for a pool is that creation of certain objects is expensive in performance terms (time, memory allocation, etc.). Rather than creating these objects over and over again, it's more efficient to create a pool of already initialized objects that can be allocated to application components, and then returned to the pool when no longer needed. Thus, the hit to performance is only incurred once, when the object is actually created.

Anyone doing performance-tuning of JDBC applications knows that creating database connections is an expensive operation. Connection pooling is a natural fit for Web applications that utilize servlets or EJBs that require many simultaneous threads to connect to databases. Figure 1 provides a more visual example. In this diagram, three application components (a servlet, JSP, and session bean) have each received a Connection object from the pool (whose current size is six).

In addition to pooling connections, the WebSphere connection manager monitors the health of a database connection. If a connection has been allocated from the pool but hasn't been used in a specified amount of time (orphan timeout), the connection manager will mark the connection as "stale" and will return it to the pool (preempt the connection).

In the event of a connectivity failure (database error, network connection dropped, etc.), the connection manager will throw a StaleConnectionException, a subclass of SQLExcep-tion, on the next JDBC operation. This allows application developers to handle failure conditions more flexibly, such as allowing the application to retry the database operation. Further, the connection manager removes all Connection objects from the pool.

WebSphere's connection manager supports a variety of databases, including DB2, Oracle, Sybase, Microsoft SQL Server, and Informix (only available in WebSphere 4.0). SQL Server and Oracle (for two-phase commit operations) require the use of the Merant SequeLink 5.1 JDBC Driver, which ships with WebSphere Application Server, Advanced Edition.

Configuration of the connection manager is simple and painless (see Figures 2 and 3). Each data source represents a connection pool to a specified database, and is linked to a single JDBC driver.

Accessing a data source programmatically requires a JNDI lookup such as:

 

Context ctx = new InitialContext();
DataSource ds = (DataSource)
ctx.lookup("jdbc/
MyDataSource");

The string, "jdbc/MyDataSource", represents the JNDI name of the datasource. In version 4.0, you may use a resource reference such as "java:comp/env/jdbc/MyDataSource", as per the J2EE specification. The DataSource object contains the methods getConnection() and getConnection(String username, String password) to allocate a Connection object from the pool.

Simply close the Connection object to return it to the pool. For example:

 

Connection con = ds.getConnection(
username, password);
...
con.close();

Now that you understand the basics of the WebSphere connection manager, let's take a look at some of the best practices.

Keep the Connection Life Span Minimized
Ideally, an application should never hold on to a Connection object beyond a single method. Nor should a single thread acquire more than one Connection. The longer one part of an application holds onto a Connection, the longer other parts cannot use that same Connection.
An application designed to acquire more than one Connection per thread and hold them for very long periods of time could easily come to a screeching halt when multiple threads collectively request more Connection objects than the maximum pool size.

The temptation here is to simply increase the size of the pool, but that will impact performance. Each Connection, whether it's in use or not, can cost over two megabytes of JVM memory. Furthermore, a higher pool size means creating more Connection objects which, as previously mentioned, is expensive. Efficient usage of fewer Connection objects is a far better design.

Additionally, when an orphaned Connection times out, the connection manager will reclaim the Connection for the pool, throwing a StaleConnectionException in the offending client on the next JDBC operation. While this is costly for the single thread, it does at least allow the Connection object to be freed. However, inside the bounds of a transaction (e.g., inside an EJB or when using UserTransaction objects), a Connection will never be orphaned. Thus, long-running transactions could keep Connection objects out of commission for a long time, and could lead to deadlock-like conditions.

While keeping connectivity life span minimized, a developer should be careful not to compromise functionality. There are times when holding a ResultSet open for an extended operation is necessary to preserve transaction isolation (row locks). Closing a ResultSet (or its creating Connection or Statement object) too early might result in your application making decisions that assumed the state of the data in the database was X when another thread has modified it to be Y.

From an application development standpoint, the best alternative is to use fewer connections in your pool, but to use them wisely. Once a method has finished its database utilization, it should free the Connection object by calling its close() method.

Cache the Datasource, Not the Connection
To take full advantage of the connection pool and make limited calls to expensive operations, an application developer should cache a datasource object when database connectivity may be utilized often. Caching the Connection has the negative effects described above and defeats the purpose of pooling.

Also, be aware that declaring Connection objects as static class variables can have unexpected and undesirable effects. For example:

 

public class MyJDBCClass {
static Connection con;
...
}

When a Connection is declared static, it's shared by all instances of the declaring class. Therefore, it's possible that two threads might attempt to utilize the same Connection object simultaneously. The results could include unexpected table locking, incorrect ResultSets, and serious data integrity problems.

Also, keep in mind that all global variables of a servlet are static by default (even if you don't declare them static). If you decide to cache a Connection, make sure not to declare it static.

Caching the datasource means creating a global instance variable in one of your classes. That class will maintain the reference to the datasource object throughout its life. When a member method needs a Connection, it creates one by calling the datasource's getConnection() method.

The advantage of caching the datasource is that you prevent multiple JNDI lookups, another expensive operation. WebSphere provides additional caching in this area, but caching it locally is advantageous.

Make Sure to Clean Up!
The WebSphere connection manager does a good job of caring for your connections, but nothing makes up for careless coding. Make sure to clean up the open database resource objects properly.

The rule of thumb when cleaning up JDBC objects is to close them in the reverse order they were created. As an example, most database read operations encompass a Connection, which creates or prepares a Statement (or PreparedStatement), which produces a ResultSet. In this case, close the ResultSet first, the Statement second, the Connection last. The order of the cleanup is important to prevent database-specific annoyances, like running out of cursors.

The best way to ensure proper cleanup is to place all cleanup logic in a finally clause of a try-catch-finally block. The code in a finally clause is guaranteed to run in all cases short of a JVM crash, so a Connection.close() in a finally clause is sure to return the Connection to the pool. Example:

 

try {
con = ds.getConnection();
ps = con.prepareStatement
("select * from mytable where mycolumn = ?");
ps.setString(1, someString);
rs = ps.executeQuery();
// do something with the ResultSet
} catch (SQLException ex) {
// handle the SQLException
} finally {
try { rs.close(); } catch(SQLException ex) {rs = null;}
try { ps.close(); } catch (SQLException ex) {ps = null;}
try { con.close(); } catch (SQLException ex) {con = null;}
}

One alternative to closing Connection objects in finally clauses is to close them in a finalize() method of an object. This approach is not recommended, as the finalize() method is only called prior to garbage collection, an asynchronous operation. Also, the only reason to postpone closing a Connection in this manner would imply that you're caching it in the object, another poor practice. Cleaning up JDBC objects correctly can prevent many unnecessary headaches.

When Good Connections Go Bad
Occasionally database connectivity breaks unexpectedly. Network cables get pulled, electricity goes out, or sometimes a database server just gets pegged. Although nothing can be done until the database is brought back online, these sorts of outages don't have to permanently bring down your application's functionality.

The WebSphere connection manager provides a convenient manner of handling temporary connectivity outages. Proper handling of the StaleConnectionException allows developers to create more fault-tolerant applications.

The StaleConnectionException is an exception that extends SQLException. This is necessary in order to follow the JDBC specification (most JDBC methods may only throw a SQLException). This is important to note because if an application is to catch a StaleConnectionException, it must be caught before catching the SQLException. The actual package name for the StaleConnectionException in version 3.5 is com.ibm.ejs.cm.portability; in 4.0 it's com.ibm.websphere.ce.cm (the 3.5 package name will work in 4.0 but is deprecated).

Just for clarification, when a Connection is marked "stale," that means that on its next operation it will throw a StaleConnectionException.

A Connection can "go bad" in any number of situations:

1.   The application attempts to acquire a Connection during a database failure or when the database has not yet been started.
2.   All of the Connection objects in the pool have gone bad due to a database failure. In this case, any new calls to getConnection() will return a stale Connection object, and all Connection objects in use would throw this exception when the application attempts to use them.
3.   The application using the Connection has previously called its close() method, but later tries to use the connection again.
4.   The application acquired a Connection but didn't use it within the orphan timeout period. In this case, the connection manager orphans (preempts) the Connection and the next time the application attempts to use it (or a JDBC resource that it created; e.g. Statement, ResultSet, etc.) a StaleConnectionException will be thrown.

Handling a StaleConnection-Exception is entirely up to the application designer. For more fault tolerance, one recommended approach is to catch the StaleConnection-Exception and retry the database operation up to a predefined number of tries. Here's an example:

 

boolean retry = false;
do {
try {
retry = false;
con = ds.getConnection();
stmt = con.createStatement();
rowsUpdated = stmt.execute("update myTable set myColumn =
5000");
} catch(com.ibm.ejs.cm.portability.StaleConnectionException sce) {
if (retryAttempts++ < MAX_RETRIES)retry = true;
} catch(SQLException sqle) {
/* handle a more severe failure and exit loop */
} finally {/*Close stmt and con */cleanupJDBCResources();
}
} while (retry);

Normal execution of this code will only execute the logic in the loop once; only in error conditions will the thread retry the operation. The connection manager could throw a StaleConnectionException on any of the operations on lines four through six. In this event, the thread will check to see if it's reached the maximum number of retries (set prior to execution), and if so retry the entire operation, starting by obtaining a new Connection object.

Handling the StaleConnectionException in this manner will allow your Web application to recover from a number of temporary database outages that would normally return an error page to your Internet customer.

Wrapping Up
The practices outlined here are recommended as best practices, and in general are good advice for building WebSphere applications. Nevertheless, none of these tips will replace a poor design; they are meant to complement a well-designed application. They're also meant to provide insight into how the connection manager functions. This information should help in future development projects as well as in maintaining your current applications.

Managing JDBC connections can be tricky, but with a good application design and by following these recommendations, you should be well on your way to delivering high-performing and scalable enterprise solutions.

Resources
Cuomo, G., A Methodology for Production Performance Tuning
Lauzon, S., and Schommer, P., Handling Database Failures
Erickson, D., Connection Manager Whitepaper
WebSphere 3.5 Handbook Redbook

These articles and whitepapers can be found online at:
www.ibm.com/software/webservers/appserv/library.html
www7b.boulder.ibm.com/wsdd and www.redbooks.ibm.com

More Stories By J. Andrew McCright

Andy McCright is a software engineer for IBM software group. His>responsibilities include development of the WebSphere Application Server>and customer engagement. Recently, he has been researching web services and how they will play into the next generation of application servers.

Comments (0)

Share your thoughts on this story.

Add your comment
You must be signed in to add a comment. Sign-in | Register

In accordance with our Comment Policy, we encourage comments that are on topic, relevant and to-the-point. We will remove comments that include profanity, personal attacks, racial slurs, threats of violence, or other inappropriate material that violates our Terms and Conditions, and will block users who make repeated violations. We ask all readers to expect diversity of opinion and to treat one another with dignity and respect.


@ThingsExpo Stories
The Internet of Things is tied together with a thin strand that is known as time. Coincidentally, at the core of nearly all data analytics is a timestamp. When working with time series data there are a few core principles that everyone should consider, especially across datasets where time is the common boundary. In his session at Internet of @ThingsExpo, Jim Scott, Director of Enterprise Strategy & Architecture at MapR Technologies, discussed single-value, geo-spatial, and log time series data. By focusing on enterprise applications and the data center, he will use OpenTSDB as an example t...
The industrial software market has treated data with the mentality of “collect everything now, worry about how to use it later.” We now find ourselves buried in data, with the pervasive connectivity of the (Industrial) Internet of Things only piling on more numbers. There’s too much data and not enough information. In his session at @ThingsExpo, Bob Gates, Global Marketing Director, GE’s Intelligent Platforms business, to discuss how realizing the power of IoT, software developers are now focused on understanding how industrial data can create intelligence for industrial operations. Imagine ...
Cultural, regulatory, environmental, political and economic (CREPE) conditions over the past decade are creating cross-industry solution spaces that require processes and technologies from both the Internet of Things (IoT), and Data Management and Analytics (DMA). These solution spaces are evolving into Sensor Analytics Ecosystems (SAE) that represent significant new opportunities for organizations of all types. Public Utilities throughout the world, providing electricity, natural gas and water, are pursuing SmartGrid initiatives that represent one of the more mature examples of SAE. We have s...
There is no doubt that Big Data is here and getting bigger every day. Building a Big Data infrastructure today is no easy task. There are an enormous number of choices for database engines and technologies. To make things even more challenging, requirements are getting more sophisticated, and the standard paradigm of supporting historical analytics queries is often just one facet of what is needed. As Big Data growth continues, organizations are demanding real-time access to data, allowing immediate and actionable interpretation of events as they happen. Another aspect concerns how to deliver ...
Scott Jenson leads a project called The Physical Web within the Chrome team at Google. Project members are working to take the scalability and openness of the web and use it to talk to the exponentially exploding range of smart devices. Nearly every company today working on the IoT comes up with the same basic solution: use my server and you'll be fine. But if we really believe there will be trillions of these devices, that just can't scale. We need a system that is open a scalable and by using the URL as a basic building block, we open this up and get the same resilience that the web enjoys.
The Internet of Things will greatly expand the opportunities for data collection and new business models driven off of that data. In her session at @ThingsExpo, Esmeralda Swartz, CMO of MetraTech, discussed how for this to be effective you not only need to have infrastructure and operational models capable of utilizing this new phenomenon, but increasingly service providers will need to convince a skeptical public to participate. Get ready to show them the money!
The Industrial Internet revolution is now underway, enabled by connected machines and billions of devices that communicate and collaborate. The massive amounts of Big Data requiring real-time analysis is flooding legacy IT systems and giving way to cloud environments that can handle the unpredictable workloads. Yet many barriers remain until we can fully realize the opportunities and benefits from the convergence of machines and devices with Big Data and the cloud, including interoperability, data security and privacy.
The 3rd International Internet of @ThingsExpo, co-located with the 16th International Cloud Expo - to be held June 9-11, 2015, at the Javits Center in New York City, NY - announces that its Call for Papers is now open. The Internet of Things (IoT) is the biggest idea since the creation of the Worldwide Web more than 20 years ago.
Cloud Expo 2014 TV commercials will feature @ThingsExpo, which was launched in June, 2014 at New York City's Javits Center as the largest 'Internet of Things' event in the world.
"People are a lot more knowledgeable about APIs now. There are two types of people who work with APIs - IT people who want to use APIs for something internal and the product managers who want to do something outside APIs for people to connect to them," explained Roberto Medrano, Executive Vice President at SOA Software, in this SYS-CON.tv interview at Cloud Expo, held Nov 4–6, 2014, at the Santa Clara Convention Center in Santa Clara, CA.
SYS-CON Media announced that Splunk, a provider of the leading software platform for real-time Operational Intelligence, has launched an ad campaign on Big Data Journal. Splunk software and cloud services enable organizations to search, monitor, analyze and visualize machine-generated big data coming from websites, applications, servers, networks, sensors and mobile devices. The ads focus on delivering ROI - how improved uptime delivered $6M in annual ROI, improving customer operations by mining large volumes of unstructured data, and how data tracking delivers uptime when it matters most.
DevOps Summit 2015 New York, co-located with the 16th International Cloud Expo - to be held June 9-11, 2015, at the Javits Center in New York City, NY - announces that it is now accepting Keynote Proposals. The widespread success of cloud computing is driving the DevOps revolution in enterprise IT. Now as never before, development teams must communicate and collaborate in a dynamic, 24/7/365 environment. There is no time to wait for long development cycles that produce software that is obsolete at launch. DevOps may be disruptive, but it is essential.
Wearable devices have come of age. The primary applications of wearables so far have been "the Quantified Self" or the tracking of one's fitness and health status. We propose the evolution of wearables into social and emotional communication devices. Our BE(tm) sensor uses light to visualize the skin conductance response. Our sensors are very inexpensive and can be massively distributed to audiences or groups of any size, in order to gauge reactions to performances, video, or any kind of presentation. In her session at @ThingsExpo, Jocelyn Scheirer, CEO & Founder of Bionolux, will discuss ho...
“With easy-to-use SDKs for Atmel’s platforms, IoT developers can now reap the benefits of realtime communication, and bypass the security pitfalls and configuration complexities that put IoT deployments at risk,” said Todd Greene, founder & CEO of PubNub. PubNub will team with Atmel at CES 2015 to launch full SDK support for Atmel’s MCU, MPU, and Wireless SoC platforms. Atmel developers now have access to PubNub’s secure Publish/Subscribe messaging with guaranteed ¼ second latencies across PubNub’s 14 global points-of-presence. PubNub delivers secure communication through firewalls, proxy ser...
We’re no longer looking to the future for the IoT wave. It’s no longer a distant dream but a reality that has arrived. It’s now time to make sure the industry is in alignment to meet the IoT growing pains – cooperate and collaborate as well as innovate. In his session at @ThingsExpo, Jim Hunter, Chief Scientist & Technology Evangelist at Greenwave Systems, will examine the key ingredients to IoT success and identify solutions to challenges the industry is facing. The deep industry expertise behind this presentation will provide attendees with a leading edge view of rapidly emerging IoT oppor...
The 3rd International Internet of @ThingsExpo, co-located with the 16th International Cloud Expo - to be held June 9-11, 2015, at the Javits Center in New York City, NY - announces that its Call for Papers is now open. The Internet of Things (IoT) is the biggest idea since the creation of the Worldwide Web more than 20 years ago.
Connected devices and the Internet of Things are getting significant momentum in 2014. In his session at Internet of @ThingsExpo, Jim Hunter, Chief Scientist & Technology Evangelist at Greenwave Systems, examined three key elements that together will drive mass adoption of the IoT before the end of 2015. The first element is the recent advent of robust open source protocols (like AllJoyn and WebRTC) that facilitate M2M communication. The second is broad availability of flexible, cost-effective storage designed to handle the massive surge in back-end data in a world where timely analytics is e...
The Internet of Things will put IT to its ultimate test by creating infinite new opportunities to digitize products and services, generate and analyze new data to improve customer satisfaction, and discover new ways to gain a competitive advantage across nearly every industry. In order to help corporate business units to capitalize on the rapidly evolving IoT opportunities, IT must stand up to a new set of challenges. In his session at @ThingsExpo, Jeff Kaplan, Managing Director of THINKstrategies, will examine why IT must finally fulfill its role in support of its SBUs or face a new round of...
"There is a natural synchronization between the business models, the IoT is there to support ,” explained Brendan O'Brien, Co-founder and Chief Architect of Aria Systems, in this SYS-CON.tv interview at the 15th International Cloud Expo®, held Nov 4–6, 2014, at the Santa Clara Convention Center in Santa Clara, CA.
The BPM world is going through some evolution or changes where traditional business process management solutions really have nowhere to go in terms of development of the road map. In this demo at 15th Cloud Expo, Kyle Hansen, Director of Professional Services at AgilePoint, shows AgilePoint’s unique approach to dealing with this market circumstance by developing a rapid application composition or development framework.