Plug-In to Denver
COLLABORATE 13 - IOUG Forum Virtual Conference
Sponsored by Cisco and NetApp
Sponsored by Cisco and NetApp
Prevalidated for Oracle
Can’t make it to COLLABORATE this year? Need to still keep up with what’s going on and get access to user driven education? Recognizing the need to provide quality group training at manageable costs, IOUG once again offers ‘Plug-in to Denver,’ a low cost of $299, virtual group training approach that offers your organization the right level of training at the right price depending on your organization’s business needs. We have two solid curriculums: High Availability, Disaster Recovery, Manageability and Performance, Scalability and Internals.
While it won’t replace an onsite experience, it does provide you access to many of the Oracle expert users, consultants and key engineers. You’ll participate by hearing the live presentation and getting an opportunity to ask questions of the speakers – as if you were in the room!
High Availability, Disaster Recovery, Manageability Curriculum
Performance, Scalability and Internals Track
Group training* that gets your team unlimited seat access to Oracle education with no associated travel and lodging expenses
- Virtual interaction with a wide-range of speakers and subject-matter experts
- Selection from about 40 hours of live, quality streaming content (selected by popular choice) that run concurrently throughout the main event
- Choice of two sessions for almost all session times from either the High Availability, Disaster Rocovery, Manageability Curriculum or Performance, Scalability and Internals Track
- Archived, on demand access to video/audio proceedings post-event
- Exclusive pass to a Webinar Q&A with Tom Kyte following the event on May 7, 2013 at Noon CST.
Tom's Top 12 Things about the Latest Generation of Database Technology
Tom will be taking a look at the Latest Generation of Database Technology and zeroing in on 12 high impact capabilities, looking at what they are and why they are relevant. Tom will also answer your questions! Visit Tom Kyte's blog to read about his latest Oracle education posts
*While group set-up provides the highest return on investment, virtual access can also be set-up at the individual level.
- Register for your group seat license.
- Once registered, you will receive a confirmation along with detailed instructions on virtual set-up.
- Anytime before the event, you and/or your employee(s) selects their session curriculum from the virtual program.**
- While you and/or your employee can access the virtual program individually, for highest return on investment, it is recommended that employers reserve a conference room or specific location prior to the event, allowing for multiple employee access.
**Note: Sessions range from 30-minute and one-hour sessions.
High Availability, Disaster Recovery, Manageability Curriculum
| Date | Time | Title and Speaker | Abstract |
| Monday, April 8 | 9:45AM - 10:45AM | Troy's Tools: Public Key Infrastructure & SSH Speaker: Troy Ligon, Nielsen Company |
Have you heard of Public Key Infrastructure and wondered what it was all about? What does it have to do with Oracle DBA's and Developers? What does it have to do with Secure SHell? In this session, Troy Ligon will be presenting another installment of his popular "Troy's Tools" series. Troy will talk you thru how Public Key Infrastructure works from a conceptual level as well as step by step how to set it up to work for you. This will be followed with an example of how to utilize this to run O/S and Database scripts securely, on both local and remote machines, without the need to hardcode passwords within the scripts. |
| 11:00AM-12:00PM | Performance Innovations with the Latest Generation of Database Technology Speaker: Amit Ganesh, Oracle |
Database performance is a key advantage that Oracle Database provides for all applications. Oracle Database has many features designed to improve performance. This session focuses on database performance innovations in the latest generation of Oracle Database technology and the benefits those features provide to applications and end users. | |
| 12:15PM-12:45PM | First PocketBook App for iphone/ipad /android : dbabuddy Speaker: Uma Lakshman, Better Life Tech |
Presenting the first pocketbook app for the DBA's and is available for the iphone and ipad. It is a free app called dbaBuddy available to download from the apple app store. The feature available now are : • Database Buddy lets you conveniently schedule and track daily tasks and upcoming tasks, as well as log your ideas for the future. • A myriad of options to assign priorities, note task time, record updates, report findings, etc. All without interrupting your work! • DbaBuddy lets you organize and save favorite links including database health monitoring reports! • You can sync dbaBuddy wirelessly with your iPhone and iPad. • Perhaps best of all, Database Buddy will keep everyone in the status loop. They are currently working on the android version currently and also looking to to enhance this one-stop app for all DBA activities. |
|
| 1:15PM-2:15PM | Manage large RAC Clusters Speaker: Tom S. Reddy, Database Administration, Inc. |
Large RAC Clusters are complex to design and install. But once they are setup, ongoing management of large RAC Clusters can be an even more daunting task. This presentation will focus on managing large RAC clusters with tips and techniques used in the real world. A combination of Oracle Enterprise Manager, custom OS scripts and other third-party tools will be discussed as potential solutions. |
|
| 2:30PM-3:30PM | Architecting Oracle database DR solution with Data Guard and SRM Speaker: Kannan Mani, VMware Inc |
Architecting Oracle database DR solution on vSphere 5 with Data Guard and SRM Abstract: This session will walk through the architecture and its components of an Oracle and VMware DR solution with Site Recovery Manager and Oracle Data Guard. This session will also focus on how to avoid issues in configuring SRM and Oracle Data Guard by following best practices , and help you be successful by implementing this solution. There will be demo on how this was setup along with DR scenarios. |
|
| 3:45PM-4:45PM | ASH: Architecture and Advanced Usage: Part 2 of 2 Speaker: John Beresniewicz, Oracle |
Active Session History (ASH) is a time-based sample of session activity in an Oracle database. It is a powerful source of information on how time is spent in the database, however misconceptions about and misuse of ASH data are common. Properly using ASH to understand and diagnose database performance issues requires understanding both its architecture and the theory of time-based sampling. This two-part presentation will give an overview of ASH architecture and the theory and methods for using ASH: Part 1 • The ASH mechanism: sampling, defaults, controls • What an ASH row represents and the multi-dimensional nature of ASH data • Estimating DB Time from ASH using SQL • Top Activity and ASH Analytics interfaces in Enterprise Manager Part 2 • The “fix-up” mechanism for TIME_WAITED and other important values • Estimating event counts and average latencies from ASH • Finding outlier events in ASH and the risk of sample errors • Visualizing DB Time and event latencies |
|
| 5:00PM-6:00PM |
Maximize Enterprise Availability with the Latest Generation of Database Technology |
Oracle Database offers a powerful set of high-availability (HA) capabilities integrated with standardized best practices for reducing both planned and unplanned downtime. Many of Oracle’s HA innovations are driven by increasingly stringent availability requirements from Oracle’s global customers, which see 24/7 demands on their own products and services continuing to multiply. Come to this session to learn how you can use the innovative HA capabilities in the latest release of Oracle Database to attain best-in-class availability for your enterprise. |
|
| Tuesday, April 9 | 8:00AM-9:00AM | Achieving a more Highly Available Environment with Disaster Recovery Speaker: Anton Els, Dbvisit Software Ltd |
What is Disaster Recovery? Do you have an Oracle database environment with important information that needs to be available no matter what? An important question you should ask yourself is; if you lose your primary database environment for a day or even an hour due to a disaster (human error, natural or any other), what would the impact to your business be? Would you lose money, clients or your reputation? In the time we live in information has become critical. In the time we live in almost everyone is making use of computers, smartphones and tablets as part of our daily lives. We are more and more dependent on information, not just socially, but also from a corporate perspective. Most companies will suffer significant losses if they lose information, even if it is just for a short time frame. So how do we protect our database environments? In this session you will be taken on a journey on what Disaster Recovery is and how using Oracle Standby database technology you can easily plan and implement a solution that can save you a lot of headache and money when disaster strikes. Areas that will be discussed include but are not limited to Oracle Data Guard, Flashback Database, Oracle Real Application Clusters, Cloud Computing as well as Standard Edition standby database technology. |
| 10:45AM-11:45AM | Internals of Active DataGuard Speaker: Saibabu Devabhaktuni, PayPal |
Active Data Guard (ADG) is one of the best features Oracle has ever introduced, but not much is known about internal workings of ADG. In this session, I talk about how redo apply and queries work in conjunction, impact of delayed block cleanout, how redo apply works in RAC, how checkpoint operation is different on the standby, what types of online changes reflected only in the controlfile, and the list of desirable enhancements to ADG. I also talk about testing portability of applications to ADG for realizing full benefits of ADG and how PayPal is using ADG. | |
| 12:00PM-12:30PM | Centralize Your Automatic Workload Repository Data for Better Analysis Speaker: Michael R Messina, ROLTA |
The Automatic Workload Repository is a powerful feature that was introduced in Oracle 10g and improved in Oracle 11g. Learn how you can utilize that power at an Enterprise level by bringing your Automatic Workload Repository data together into a central repository to compare reports across databases in your Enterprise; keeping snapshot data for longer periods without taking space in your live databases; thus adding to the flexibility and power of the Automatic Workload Repository. This session will give you some ideas of how you might leverage this feature in your Enterprise. Presentation will cover the central AWR components and processes then a demonstration of central AWR and an APEX interface will follow to demon straight capabilities of the central AWR. | |
| 2:00PM-3:00PM | Oracle 11gR2 Active Duplication. Cloning a production Instance to development. Speaker: Howard Hackworth, Eli Lilly |
Oracle 11g offers several methods for database duplication, including duplicating to the same server or duplicating to an alternate or development platform for use as development instance. This talk will provide a real-life step-by-step example of how I use Oracle Active Duplication in 11gR2 to duplicate or clone my production instance to a development instance on a development server, thus providing developers with an updated version of production for testing. I will discuss the different types of duplication available in 11g and then dive into a step-by-step approach for duplicating your production database to a development instance on a development server. Topics include assessing requirements, choice of duplication type, specific RMAN examples, and steps for reconfiguring OEM for the cloned database. | |
| 4:30PM-5:30PM | Oracle Enterprise Manager Deployment Best Practices Speaker: Werner De Gruyter, Oracle |
Everything you’ve always wanted to know but were afraid to ask: this session addresses some of the critical issues associated with designing for Oracle Enterprise Manager deployment, sizing the infrastructure for short- and long-term needs, agent deployment and management best practices, patching and monitoring best practices with Oracle Enterprise Manager Cloud Control’s Oracle Management Service component, securing your environment, and how to implement with high availability and disaster recovery in mind. This is a must-attend session for anyone just embarking on a new deployment or planning for upgrades or future expansion. | |
| Wednesday, April 10 | 8:15AM-9:15AM | RMAN: New Features Speaker: Michael Abbey, Pythian |
New features bundled with Recovery Manager have been progressively making a strong product even stronger. Oracle has attended to some useability enhancements in particular which address the difficulty of backing up and restoring huge information repositories. Reduced overhead for operating system snapshots as well as the ability to breakdown and parallelize the workload of backup and recovery stand out. Rman has always been a candidate for table-level recovery which is now bundled with the product. This presentation will walk the attendees through a table-level recovery activity. |
| 9:30AM-10:30AM | You most probably don't need RMAN catalog database Speaker: Yury Velikanov, Pythian |
The title of this session is on purpose thought provoking. The author is an experience Oracle DBA in Oracle backup & recovery area. During the presentation he will go through top reasons why you may need to implement RMAN catalog database and give you additional ideas on how you can improve your backups leveraging additional benefits provided by RMAN catalog database. The author will explain in what cases and why you may not need the catalog database. You will go away with a clear understanding on how to benefit from RMAN catalog database and when it may be optional. This is another presentation from author's popular RMAN papers. | |
| 11:00AM-12:00AM | DRA - Amazing toolkit to reduce the MTTR Speaker: Nallasivan Inoothumuthu, Verizon Wireless |
This technical session focuses on how to automatically detect, analyze, and resolve various data failures . This session provides a walkthrough using real-life scripts in detecting database/data recovery issues along with determining the scope of the issue. The presentation leverages several of RDA commands to dig into to uncover the failures with various ways to recover from them. The takeaway from this session is: “How a DBA who has no skills with RMAN can comfortably recover the database from many failures." | |
| 12:15PM-12:45PM | Build a RAC Database for Free with VirtualBox - A Step by Step Guide Speaker: Chris Ostrowski, Avout |
This presentation will walk attendees step by step through the process of building a RAC environment for free. Oracle's Real Application Clusters (RAC) and Grid Infrastructure are technologies that provide organizations with high availability for their Oracle database environments. For many Oracle professionals however, it can be challenging to learn RAC - while proactive DBAs can easily download and use any of Oracle's products for learning purposes, having access to or creating a RAC environment can be a serious limitation for many. In this presentation, I'll show you, step-by-step, how I created a RAC environment using nothing but tools freely available via the Internet. |
|
| 1:00PM-2:00PM | Maximum Application Availability with the next generation Oracle database technologies Speaker: Kuassi Mensah, Oracle |
Prior to the next generation Oracle database technologies, Fast Connection Failover and TAF allow failing over connections and queries but do not guarantee transaction outcome, failing over transactions, or masking outages. This session covers new Oracle database availability including Transaction Guard for checking and guaranteeing transaction outcome, Application Continuity for attempting to mask outages by failing over connections and replaying session contexts, queries, transaction (DML), DDL, and so on) by failing over and session context replay handling outages. Other new HA features include recoverable errors, fast dead connection detection and adaptive connection retries. Application developers and DBAs will learn how to leverage and enable these features. |
|
| 3:00PM-4:00PM | How to minimize the brownout, due to RAC reconfiguration. Speaker: Amit Das, PayPal |
Whenver any instance leave or join in the cluster, it will cause session spike. How you can minimize the session spike? RAC's best practice and beyonds. |
|
| 4:15PM-5:15PM | Virtualized Oracle Stretched RAC Cluster using VMware vSphere & EMC VPLEX Speaker: Jeff Browning |
This presentation will present a solution for creating a virtualized stretched RAC cluster. This cluster will provide much lower latency, improved availability, automated failover, simplified administration, and many other benefits over creating a stretched RAC cluster using conventional means (ASM). There are three basic layers to this solution and the presentation will be split up accordingly. First, there is the Oracle side, which will be presented by Charles Kim (one of the foremost experts on Oracle). Then will come the virtualization layer, presented by George Trujillo. Finally, will come the storage layer (including the EMC VPLEX, which is the heart of stretched RAC), presented by EMC's Jeff Browning. |
|
| Thursday, April 11 | 8:30AM-9:30AM | Tech refresh of existing system with ZERO downtime using RAC, ASM Technology Speaker: Amit Das, PayPal |
We do tech refresh almos every 2 years as our business is growing too fast. But we can not take downtime for this. We have upgraded our AIX-p770 to AIX-p795 with ZERO down time. We did the same from Solaris T3 to T4 upgrade. In the upgrade process, even we have upgraded our private interconnect from 1GigE to 4X10GigE interconnect. |
| 9:45AM-10:45AM | Near Zero Downtime Oracle RAC Migration with Oracle Automatic Storage Management and Oracle Enterprise Manager 12c Speaker: Wendy Chen, Dell |
Performing a seamless migration to move Oracle Database to new hardware platforms and new operating systems can be challenging. This session shares Dell’s experience in a recent Oracle Database migration project to move Oracle Real Application Clusters (RAC) databases to new storage systems, new servers, and new operating systems running on the database servers. It provides best-practice migration methodologies for moving Oracle RAC databases into a new infrastructure with near zero downtime impact on the database availability. Key technologies involved in the migration included Oracle Automatic Storage Management disk rebalance and Oracle Enterprise Manager 12c. | |
| 11:00AM-12:00PM |
Zero Downtime Migrations using Oracle GoldenGate |
This presentation explains minimum downtime migrations between various Oracle version and platform combinations with Oracle GoldenGate. |
Performance, Scalability and InternalsTrack
| Date | Time | Title and Speaker | Abstract |
| Monday, April 8 | 9:45AM - 10:45AM | Optimizing Internal Serialization Control (latches and mutexs) Speaker: Craig Shallahamer, OraPub, Inc. |
Memory serialization control is core to Oracle database operations. When serialization is used, how it works, how we can influence its operation, and how we diagnose problems is one of the most fascinating topics in Oracle internals and performance optimization. But it's a complex situation. To get the upper hand on serialization control this presentation will explore the lock (a little), the latch (a lot), and the mutex (a whole lot). Topics include performance diagnosis, how Oracle implements latches and mutexes, and the related internal algorithms. Special attention is given the library cache mutex operations. This is a very practical yet deep internals presentation, filled with amazing discoveries about how Oracle works. |
| 11:00AM-12:00PM |
ASH: Architecture and Advanced Usage: Part 1 of 2 |
Active Session History (ASH) is a time-based sample of session activity in an Oracle database. It is a powerful source of information on how time is spent in the database, however misconceptions about and misuse of ASH data are common. Properly using ASH to understand and diagnose database performance issues requires understanding both its architecture and the theory of time-based sampling. This two-part presentation will give an overview of ASH architecture and the theory and methods for using ASH: |
|
| 12:15PM-12:45PM |
An introduction to Self Service (Troubleshooting) Tools forORA-600/7445/4031/4030 errors |
An introduction to the new web-based Troubleshooting Tools for ORA-600, ORA-7445, ORA-4031 and ORA-4030 errors that enables |
|
| 1:15PM-2:15PM | "Where did my CPU go?" – Monitoring & Capacity Planning Adventures on a Consolidated Environment Speaker: Karl Arao, Enkitec |
This session will focus on CPU monitoring and capacity planning. We will highlight scenarios that are typically encountered on a massively consolidated environment, where, let's say you have 30+ databases and you want to know how much CPU cores they are using at a particular time interval. Before touching on the cool tricks, a deep dive into important CPU metrics is a MUST. In Oracle world the CPU is not just the "Green Thing" in the Enterprise Manager. It's actually much more than that. We will discuss the usual CPU monitoring tools in OEM, visualization enhancements that can be made by AWR analytics, and how these can be applied to critical capacity planning scenarios. | |
| 2:30PM-3:30PM | NOW What's Up with dbms_stats? Speaker: Terry Sutton, Database Specialists, Inc. |
Having accurate optimizer statistics for your database is more important than ever. Handling larger volumes of data and benefitting from the ever-increasing number of choices in execution plans generated by the Oracle Optimizer require that the optimizer knows as much as possible about your data. The only reasonable way to gather the statistics is with the dbms_stats package. But there are many options to dbms_stats, and your choice of which options to use can dramatically affect your results, both in accuracy of statistics and performance of the statistics gathering operation itself. A few years back we examined dbms_stats in Oracle 9 and 10.1. We update this information for Oracle 11.2 (and the next generation of the Oracle database, if it is released soon enough). In this presentation we will discuss the effects of the various choices. The focus here will be on actual experience, measured performance, and detailed examples— not just on what the documentation says. Those attending the session should have some experience with gathering optimizer statistics. |
|
| 3:45PM-4:45PM | SQL Tuning made easier with SQLTXPLAIN (SQLT) Speaker: Carlos Sierra, Oracle America, Inc |
SQL Tuning is a daunting task. Too many things affect the Cost-based Optimizer (CBO) when deciding on an Execution Plan. CBO statistics, parameters, bind variables, their peeked values, histograms and a few more are common contributors. The list of areas to analyze keeps growing. Over the past few years Oracle has been using SQLTXPLAIN (SQLT) as a systematic way to collect all the information pertinent to a poorly performing SQL statement and its environment. With a consistent view of this environment, an expert on SQL Tuning can perform a more diligent task, focusing more in the analysis and less in the information gathering. This tool could also be used by an experienced DBA making his life easier, at least when it comes to SQL Tuning. | |
| 5:00PM-6:00PM | PHYSICAL ORDERING OF DATA: IS IT EVER USEFUL? Speaker: Mark W Farnham, Rightsizing, Inc. |
Certain operations, some small like lookup lists, and some large like batch jobs and reports, can perform better if the physical order of the data matches the predominant order of use. When is it worthwhile to attempt this ordering? How does ASSM affect the ability to physically order data? An expensive mission to periodically physically order all the data you have in the best possible order for the queries you expect to run on the data is silly. But if you verify that a one time complete or periodic partial physical ordering of some of your data will have a reasonable cost that is nearly certain to be exceeded many times over by the reduction in cost of future projected queries against a constrained resource, then it is not silly. |
|
| Tuesday, April 9 | 8:00AM-9:00AM | In-Memory Database Processing with the Latest Generation of Database Technology Speaker: Amit Ganesh, Oracle |
Every DBA knows that online transaction processing (OLTP) workloads run mostly in-memory within the database buffer cache. However, not every DBA recognizes how Oracle Database provides a tremendous breadth of in-memory optimizations beyond simple caching of database blocks. Databases today have large volumes of memory available—sometimes ranging to terabytes—and Oracle Database has been optimized to make full use of these memory resources. In this session, learn how Oracle Database now leverages column-based data structures, optimized in-memory scans, efficient data compression, result set caching, and large-scale in-memory parallel processing across clusters to deliver increasing levels of performance for all database workloads by making more-efficient use of available memory. |
| 10:45AM-11:45AM | Resolving Cache Buffer Chain Latch Contention Speaker: Craig Shallahamer, OraPub, Inc. | The numbers are intimidating; perhaps trillions within a single hour. When a server process needs to access a buffer it must first access the cache buffer chain (CBC) structure to determine if the buffer is in the buffer cache. While serialization control protects the CBC structure, given high concurrency and just the right workload mix, performance can be a problem. Through amazing visualizations and entertaining role playing, you'll learn the relevant Oracle internals, quantitative methodical problem diagnosis, and solutions related to the three most common CBC struggles: a hot index root block, extremely high buffer cache chain concurrency, and buffer cloning. | |
| 2:00PM-3:00PM | Optimizing Oracle database with SSD Speaker: Guy Anthony Harrison, Dell software |
Solid State Disk (SSD) is rapidly becoming a mainstream technology and an almost indispensable ingredient in a high performance Oracle database. While SSD is often too expensive to replace all your database disk, judicial use of SSD in combination with magnetic disk can be a very cost effective way of accelerating Oracle database performance. This presentation will consider how best to use SSD in an Oracle 11g database to get the most bang for your SSD buck. You'll see examples of configuring SSD: • as part of the database flash cache • to improve IO for selected tables or partitions • as part of a temporary tablespace to improve disk sort performance • to accelerate redo log IO • inside an Exadata system |
|
| 4:30PM-5:30PM | How Oracle Secretly Changed Multiblock Reads Speaker: Frits Hoogland, VX Company |
This presentation is about how the Oracle database implements multiblock reads on Linux systems. Whilst this may look like a simple and easy to understand topic (the system fetches multiple blocks conforming db_file_multiblock_read_count blocks instead of one, right?), in reality it isn't. The description of the former line is mostly true for version 10 non-PQ multiblock reads, but with Oracle version 11 it has changed. Oracle silently introduced true asynchronous reads with version 11, called 'adaptive direct path reads', which happen under specific circumstances. This session outlines these circumstances. One of the most eye-catching features is reading blocks to the PGA, which makes the reads non-shared, which is different from the traditional reading to buffer cache/SGA. |
|
| Wednesday, April 10 | 8:15AM-9:15AM | Compression in Oracle Database Speaker: Tim Gorman, Evergreen Database Technologies, Inc. |
Partitioning and compression are technologies that have been in the Oracle database for a long time. The EXCHANGE PARTITION command is the key to large-scale and scalable data loading in a data warehouse, and a key prerequisite for the effective use of compression. But now there are many different types of compression available in Oracle database, and which are most appropriate for which purpose? This presentation will discuss all the different types of compression in Oracle database (i.e. basic/advanced de-duplication compression and hybrid-columnar compression), as well as some older techniques for fitting more data into less space. Then, the presentation will discuss the ways to take advantage of each type of compression most effectively. |
| 9:30AM-10:30AM | SQL "Pagination" Pattern: Designing Efficient Top-N and Pagination Queries Speaker: Maxym Kharchenko, abc |
"Pagination" (of which Top-N is a special case) is a very common query technique. It deals with extracting a limited number of "most interesting" records from a potentially large qualifying result set. While pagination requests seem simple (and the next generation of the Oracle database made them even simpler), executing them efficiently requires some ground work in both schema design and query coding. In this presentation we will take on a few real life pagination examples and use several techniques (i.e. "perfect indexes", "restart tokens") to make them execute efficiently. We will also discuss a number of pitfalls and ways to avoid them. |
|
| 11:00AM-12:00AM | Improvements In Oracle 11gr2’s Statistics and Its Implementation with Partitioned Tables Speaker: Tony Jambu, Wizard Consulting Pty Ltd |
Oracle 11g has been in production since July 2007 but to-date, its statistics gathering procedure DBMS_STATS, is not well understood nor has it gained the confidence of users to accept its default parameters. This paper will demonstrate with examples and demos how the DBMS_STATS statistics gathering has improved significantly and is gathering more accurate information while being more efficient and faster. In addition, one of the best and not so well known feature is its Incremental Stats gathering with (sub)-partitioned tables. Table statistics gathering that used to take hours can now be completed in minutes. Numerous users have voiced how this feature does not work as documented. We will show why this may be so and how to achieve this correctly. |
|
| 1:00PM-2:00PM | How to improve SQL Performance with new SQL Health Check Tool Speaker: Carlos Sierra, Oracle America, Inc |
When your report to Oracle a SQL with poor performance, most probably you will be asked to provide a SQLTXPLAIN (SQLT) report. This SQLT tool is used by Oracle to diagnose SQL statements performing poorly. The problem that you may face is that SQLT requires to be installed before being used. In some Production systems this means scheduling an activity for a future maintenance window. As a workaround to this potential delay, you can benefit of the new SQL Health-Check (SQLHC) tool. SQLHC does not require to create any objects into the database, so it can be used easily on a Production environment. It performs dozens of health checks around the SQL statement being analyzed. It is not uncommon that acting upon the health-checks actually fixes the poor performance of your SQL. During this session you will learn how to use SQLHC and what type of health-checks it performs. As SQLHC matures it has gone through several extensions. It does not compete with the value provided by SQLT, but the gap between the two is closing. SQLHC provides now a lot more diagnostics details than when the tool was first released. Still it keeps its fundamental difference with SQLT: SQLHC requires to install nothing on the database! |
|
| 3:00PM-4:00PM | Demystify Global Temporary Tables: All About Resource Management and Optimizer Statistics Speaker: David Chen, Concuity, Advisory Board Company |
It’s no secret that proper usage of Global Temporary Tables (GTT’s) can result in tremendous performance improvement and dramatic resource usage reduction. This is because data for a GTT are stored in temporary segments and redo logs are not generated. Undo creation is minimized as well. Generally GTT’s are good for applications where there are a lot of transient DMLs, or many data transactions are in a “staging” state. However, GTT’s cannot be analyzed. And, even though the data are not shared, any statistics set for them are shared across sessions. As such, optimizing GTT’s can be tricky. Also, when a GTT grows to a certain size, query performance can deteriorate drastically. • So what is the threshold? • What are the things to be aware of when implementing a GTT? • How do you collect/set/share statistics for these tables among concurrent sessions? • Is there any way to simulate histogram for GTT to improvement performance of complicated queries involving GTT? In this session, we’ll shine some light on the mysteries around the use of GTT’s, statistics and resource management. We’ll walk through a live demo based on data analysis of an industry leading healthcare revenue recovery calculation engine. Best practices for using GTT’s will be shown, as well as statistics gathering and setting for feeding to the Oracle optimizer. The results of these techniques result in optimized performance and minimized resource usage. We’ll do a comparison of GTT’s versus permanent tables, and transaction versus session GTT’s, and evaluate parallel session GTT statistics and resource interaction which will provide a complete picture about the benefits of GTT’s. If you are using GTT’s, you cannot afford to miss it! |
|
| 4:15PM-5:15PM | All About Database Services Speaker: David Aihe, SAIC |
Database services have been around for a while, however, most DBAs typically do not know the full benefits of services. Some DBAs think replacing the SID variable with SERVICES in their tnsnames.ora file is all they need to do. This paper presents a detailed analysis on the use, configuration and benefits of oracle services in a database. Services will be discussed in the context of non-RAC and RAC databases. | |
| Thursday, April 11 | 8:30AM-9:30AM | NFS Tuning for Oracle Speaker: Kyle Hailey |
NFS throughput has passed fiber channel and that gap over fiber channel is expected to keep increasing. NFS is easier and cheaper to install, configure and maintain than fiber channel and includes robust error handling such as check summing and retransmissions. On the other hand NFS has a reputation as being slow. Learn if NFS is truly slow or if this is a myth and learn how NFS can be optimally configured and how to analyze for NFS bottleneck. NFS bottlenecks can be hard to identify using standard tools, but armed with the ultra powerful analysis tool DTrace on Solaris we can get a surprisingly rich picture NFS communication. |
| 9:45AM-10:45AM |
RDBMS Forensics: Troubleshotting Using ASH |
Arthur C. Clarke wrote that "any sufficiently advanced technology is indistinguishable from magic." This has been proven true in many technologies, and it makes troubleshooting of that technology seem like art. But it is not art, it is process, and the intent of this presentation is to use a case study (i.e. resolving ORA-03136 "connection lost" error) to demonstrate that process, and also to show how components such as AWR and ASH, typically thought of as "performance tuning" tools, are also useful diagnostic tools. Attendees to this presentation will learn a forensic, empirical approach to trouble-shooting and also see a demonstration of a complex troubleshooting task following a non-intuitive but empirically sound resolution. | |
| 11:00AM-12:00PM | Queues, Pools and Caches - The Right Way to Scale OLTP Speaker: Gwen Shapira, Pythian |
Transaction processing systems are generally considered easier to scale than data warehouses. Relational databases were designed for this type of workload, and there are no esoteric hardware requirements. Mostly, it is just matter of normalizing to the right degree and getting the indexes right. The major challenge in these systems is their extreme concurrency, which means that small temporary slowdowns can escalate to major issues very quickly. In this presentation Gwen Shapira will explain how application developers and DBAs can work together to built a scalable and stable OLTP system - using application queues, connection pools and strategic use of caches in different layers of the system. |
*** Please note that all sessions are subject to change. For the most up to date schedule please view the session scheduler.