The 5 Core Issues Facing Database Administrators

Every data centre manager, be it at a Fortune 500 or a small startup, must face five core issues during his or her database management. Two fundamental keys of databases themselves are exacerbated by three other forces that require attention in order to have a robust, functional, and powerful database system.

The nature of DBA is simple: ‘If you’re doing your job well, nobody will notice. If you screw it up, everyone will notice.’

=== DB Fundamental: Accessibility

To be useful, a database must be accessible. Otherwise it is just a store of information that no one can retrieve. In database management terms, the value of data is measured by how much it costs (in dollars and/or machine function/processor time) to store and retrieve it. The less it requires to get to the data, the cheaper the data is by these metrics.

This means that the easier it is to find data, the cheaper that data is to the overall scheme. For example, in 2005, Cisco estimates that it spent $25.2 million on data retrieval just for regulatory compliance purposes. After upgrading and improving their processes, they cut that cost by 97% to $1.8 million the next year. Not insignificant.

So accessibility is perhaps the top priority for nearly every DBA and her team.

=== DB Fundamental: Change Management

Two types of change management occur in a database. The first is management of the data’s changes themselves – how they are accessed, changed by users, and stored. The second is how the business around the database changes, affecting the database itself.

The first DBA concern is user management and this ties in closely with security (below). Most DBAs realize the importance of keeping date-stamped backups of original data so that if something is changed and the changes must be reversed, they can be. That’s the basics of change management in data.

In business, changes are constant and for the DBA, this means new hardware, changes in software, upgrades to client machines, and more. All of these can affect the DB, so the administrator must keep a fluid system capable of growth or alteration. Managing the lifecycle of your hardware, for instance, can mean 30% in savings over a year thanks to lowered replacement and disposal costs.

=== DB Forces: Compliance

Likely the most important (and hated) aspect of database management is compliance. There are whole volumes of legal books dedicated to document and data storage requirements as well as retrieval needs for those legal documents the company stores. Literally every department of a business will have compliance requirements for the DB. Often these must be preserved for long periods of time as well.

Balancing the data storage needs of the company with its retrieval requirements is often the most demanding thing a DBA does.

=== DB Forces: Security

If the news is any measure, data theft and losses are on the rise. While this has been a core issue for DBAs since the days of DBA 101 classes in school, it’s becoming a forefront subject as well as the impetus for many changes happening in the business today.

=== DB Forces: Disaster Recovery

‘Backup once, then backup regularly’ is the mantra of all IT workers. In DBA it’s said twice. Nothing is more disastrous for a company than having its database lost and irrecoverable. The ability to quickly restore lost data, change access accounts and passwords, etc. are paramount to both security and asset safety.

These five issues can be considered the core of what all DBAs face and should be what they focus on the most.

Data Consolidation Techniques

As data increases in the company’s database, the need for data to be consolidated is a must in order to manage it effectively and utilize it for business operations. Data consolidation is getting data from multiple locations and sources and integrating them into a single database to be used in the company. Consolidation is an important component in data integration modules that comprise data propagation and federation.

Data propagation deals with duplicating information from different sources and locations while data federation deals with unifying the source information virtually. When data is integrated into a single database it allows for quicker access and better control. Managing data is now more effective and efficient. Data consolidation is done with the use of two different technologies and these are the ELT and ETL.

ELT stands for Extract, Load, and Transform. This is where the systems transform a volume of data after it is loaded in a database. After the loading process is done, it is then transformed and then delivered to different tables that can be access by authorized users. This technology is also called pull systems because it is performed on-demand by any individual. This allows also the users to transform and publish data after it is loaded in the database.

On the other hand, ETL stands for Extract, Transform, and Load. This is another data consolidation technique where it extracts information from multiple resources, transforms it into the standard rules and then loads it afterwards in the target systems with specified formats. It is quite different from ELT, because data is being transformed first before loading process takes place. Transformation takes place in the form of reformatting, standardizing and streamlining it to other data manipulation rules set by the company.

The extraction process is the first stage in any data consolidation techniques. Extraction may take place from high volume to multiple data sources or maybe from relational to object databases and other documents. This may also delivers both unstructured and structured data. The next technique is the transformation process that varies from data consolidation technique that is available. This may also ranges from single to complex operations. This allows also to deliver timely and relevant information that are used by the management team in their decision making process. Data is customized and tailored to what the company really needs. And the last process is the loading where it transfers and delivers data from one location to any target application. The loading process differs in both techniques because in ELT data loaded is unprocessed while in ETL data is loaded after it is processed.

Data consolidation is done with two different techniques. However, both of these techniques aim to integrate all the necessary data and information from different sources to a single database for effective management of data.

How Does A Co-Working Space Score Over Any Office Space?

With the boom in e-commerce, many start-ups, before actually finding a professional office space and building a team, used to work from their bedrooms, kitchen, garage, basement and living rooms. An idea can germinate at any place, but it requires a perfect ambience and nourishment for execution.

For this execution to come into existence, the advent of co-working spaces became a trend that is being followed by the world today. This has brought about a major shift from owing a conventional office space that ties up the capital.

Rather than considering to work from home, many start-ups and business owners are now finding a shared office space as the most desired option. To your amusement, it is not just the start-ups or entrepreneurs who are opting for such a set-up, even, major MNCs and corporate houses consider using this option for their back-office work.

Viral trend

This trend is becoming viral just because, one, it is cost-effective and two, it gives one a perfect professional set-up wherein one can work efficiently. The efficiency of doing the same work increases many times when one is provided with a professional environment to work. It helps in sustaining interest in your business which otherwise could have been relegated to doing some other menial work in any corner of your home.

Building confident entrepreneurs

Those who are using such co-working spaces claim to beat conventional working space hands down. The level of confidence among those who considered using any other office space, be it a home or a conventional one, has risen considerably. They are now more confident in their dealings and reflect a professional approach towards their work.

Highly flexible

A shared office space provides them perfect flexibility to ply along with increasing their network. There are other bright minds who are also working towards making their business a success. Such an atmosphere induces buoyancy in your work and also increases your chances of getting noticed.

Amidst these co working spaces, nowadays, there are a lot of possibilities of expansion. When it is about being flexible, a lot of options are available for that take off stage when you are ready to build your team for a perfect execution. Business centres provide a lot of cushion to any hiccup you might experience during your business.

Pick and choose

The best point in opting for a shared office space is that you can actually pick and choose a convenient location that may be near your client base or your home. Such an elasticity increases the quotient of happiness that in turn reflects in your ease of doing business.

Feel connected instead of isolated

While working from home leaves you less motivated and isolated, opting to work in a shared office environment lets you breed new ideas and innovate. It helps you stay connected and be a part of a community. Moreover, you are also surrounded by technically advanced options on your desk to be connected to the world.

Access to professionally ethical space: Meeting rooms

Working from home might make you happy but meeting your clients at home will never make your business prosper. This might lead you to meet them in public places like restaurants and hotels which is professionally unethical. A shared office space lets you reflect a perfect professional image on your client with options of hiring a meeting room.

Earn that credibility of a professional office space even before you happen to meet your client. Preparing things in advance lets you play a safe and sure game. With various technically advanced options like video conferencing for a client who needs to access his headquarter situated abroad while in a meeting with you helps you score points already.

Why deliver your first ever business presentation in a background that is full of noise and unwanted glances. Doing it on a projector in a perfect professional set-up increases your chances of getting noticed and heard. Such meeting rooms are available with latest AV equipment to make your first impression a lasting one.

Metropolitan cities in India are not averse to this concept. In fact, there has been a lot of boom in this business of offering shared office space in Delhi, Mumbai, Kolkata, Chennai, Hyderabad, Pune, Bengaluru etc. Their business idea resonated with a demand that was untapped and needed an outlet for exposure. They worked out on how to keep the expenses of a new business minimum and yet get the best out of it. Here, it came out to be a successful business model for many others to emulate: co-working space provider.

Disaster Recovery Invocation Procedures

The following procedure illustrates at high level the first 24 hours following disaster invocation. This procedure is based on a “warm” recovery service.

Following a disaster, clearly defined steps/actions need to be taken to enable business continuity. During the first 24 hours these steps will fall into the following categories.

Initial Assessment

Timescales – Immediately (T + 0)

Following a disaster situation the first step that must be taken is to assess the current situation. This will be carried out by the Disaster Co-ordinator, who will decide if the Disaster Management Team needs to be assembled. The team will need access to a Disaster Command Facility, if the primary location is not accessible for any reason. The Disaster Management Team and Command Centre should be detailed, along with relevant phone/mobile numbers and directions in the Business Continuity Plan.

The relevant emergency services should have already been notified of the situation. The Disaster Management Team would act as the main focal point for the emergency services.

It may be necessary to make a pre-invocation call to put the Disaster Recovery service on standby, thereby reducing the response time should the service be formally invoked.

Disaster Management Meeting

Timescales – within 1 hour (T + 1 hour)

If it is necessary to call a formal Disaster meeting, this should happen within 1 hour of the event. It may not be possible to get all members of the team together in these timescales, therefore all essential members should be agreed upon and documented in the plan.

The Disaster Management Team’s main role would be to:

­ Define the problem

­ Define the extent of the disruption

­ Determine the likely impact on your business

­ Estimate outage length (where possible)

­ Invoke Disaster Recovery service if applicable

­ Formally set up Disaster Command centre

­ Agree team’s objectives for next three hours

­ Agree formal verbal report for senior management

­ Agree on staffing levels needed at the present time

­ Send non-essential staff home (if during office hours)

­ Contact non-essential staff at home (if out of hours)

­ Call in additional staff (if out of hours)

­ Set up next meeting for T + 4 hours

Disaster Review Meeting

Timescales – within 2 hours (T + 2 hours)

At this stage you should have a much more detailed understanding of the situation. This will enable a full written report to be produced for senior management.

The Disaster Management Team will have by this time:

­ Invoked the disaster Recover Service (if applicable)

­ Set up a temporary Disaster Command centre

­ Mobilise essential staff members

If applicable the warm standby (Disaster Recovery) services should be available by this time to start configuration of the standby systems.

Configuration of Standby Equipment

Timescales – within 2 hours of invocation (T + 4 hours)

Warm Disaster Recovery configurations are normally scheduled to be available within 2 hours of invocation. By this time the site should be ready to receive the equipment. Power and Communications should be enabled and facilities for the essential staff should be available. Additional equipment needing to be purchased may arrive some time after this. The backup media will also have arrived onsite.

Restoration of Data and Testing

Timescales – within 20 hours of invocation (T + 22 hours)

Up to 8 hours may be required to restore and test the system. Comprehensive user acceptance test (UAT) procedures should be documented in your Disaster Recovery Plan to ensure the systems are fully operational before they are announced to be live to the end user.

Systems available to end users

Timescales – within 22 hours of invocation (T + 24 hours)

At this stage you should be able to resume some (or all) of your business activities (depending on the scope of the disaster). It is critical at this stage to plan for full business restoral. These steps should include:

­ Interim requirement such as larger temporary accommodation

­ Refurbishment of damaged offices (if applicable)

­ Identification of new premises (if applicable)

­ Replacement of damaged equipment

A full Business Resumption plan should also be produced, detailing the transition from the standby facility to permanent offices.

Visit EMAContinuity @ http://emacontinuity.com

What Is the Difference Between Hot, Warm and Cold Disaster Recovery?

When it comes to implementing your business continuity plan what strategy do you adopt for the disaster recovery element? (for a description on the difference between Disaster Recovery and Business Continuity please see my article on Disaster Recovery or Business Continuity?).

You may have heard the terms hot, cold and warm recovery, but what do they mean, and what are the advantages and disadvantages of each service?

Hot Standby

Hot standby is normally available to the users within minutes of a disaster situation. This level of service is achieved by total duplication of the computer systems covered (hardware, software and data). There will also be a requirement for a resilient network connection into the Hot Site.

Benefits – Available immediately; dedicated to (customer).

Disadvantages – Cost; Complexity, management.

Warm Standby

Warm standby is normally available to the users within hours of a disaster situation. This is by far the most common type of service utilised by for I.T. disaster recovery, and typical recovery times range from 8 hours to 24 hours (dependent on complexity, location and data volumes).

The service can be delivered from a remote recovery centre, or alternatively, delivered to site in the event of a disaster. Depending on the equipment involved the configuration may be installed within an existing facility or a mobile recovery unit.

It should be noted that whilst the Hot standby option is normally dedicated to one customer, Warm standby is delivered on a subscription basis. Industry standards are between ten and twenty five subscribers per configuration. Availability is therefore not guaranteed in the event of a disaster. Testing is also normally to a predefined number of days P.A.

Benefits – Lower cost; reasonable availability.

Disadvantages – Availability; recovery timescales are longer; limited testing available; only available for a limited period following a disaster.

Cold Standby

Cold standby is the provision of computer and people facilities that are made available to the client within a few hours of the incident. Unless the service is backed up by a contract to supply the necessary computer equipment, the recovery period is likely to be several days. It is not unusual for Warm and Cold standby services to be combined, giving a very flexible approach to recovery.

Fully serviced office space is also available on a subscription basis. These are usually equipped with PCs, servers, printing facility and a network infrastructure. These would be described as Business Recovery Centres, and could also incorporate Cold space for central systems.

Benefits – Lower cost; large amount of available space (can accommodate large systems). Business recovery Centres can accommodate several hundred people.

Disadvantages – Availability; recovery timescales are longer; limited testing available; only available for a limited period following a disaster; additional recovery services needed.

EMA Continuity

Diagnosing And Troubleshooting Computer Hardware

What is Hardware?

Are these terms familiar to you? Monitor, RAM, CD drive, CPU, graphic cards are all hardware, or more accurately, computer hardware. These components make up a computer, working together with computer software to make a computer work.

Simply put, hardware is the tangible parts of your computer, parts you can touch, feel and so on.

Hardware Failures

As mentioned before, hardware and software work together to form a fully functional system, theoretically. However, rarely do you get a fully functional system all the time. There will almost certainly be malfunction of hardware, be it within the electronic circuits or even the whole component itself. Sometimes, the originating factor of the hardware failure is not the components of the system itself, but due to outside factors such as environmental disasters like fire, earthquakes and lightning storm.

Recovery of failed hardware components is not a big problem by itself. It is basically identifying and replacing the problematic component. However, hardware failures are most deadly, when it affects daily routine and affecting critical personal or business data. This is especially true of the most important component of a computer system when it comes to storage of data, the hard disk.

The following is a list of common hardware failures:

  • RAM Failures
  • Power Connector
  • Hard Disk
  • Overheating
  • LCD Failures
  • Motherboard
  • USB Box

RAM Failures

Bad RAM is somehow harder to diagnose as similar symptoms may be caused by software problems, other hardware problems or even motherboard failure. However if you experience any of these symptoms, users should check for bad RAM before attempting any other troubleshooting.

Symptoms:

  • Windows doesn’t start showing different error messages each time.
  • Windows crashes (blue screen) or freezes frequently.
  • Windows crashes as soon as you try to start a program.
  • Unexplained random crashes and freezes without error messages.

Power Adaptor

The common weak spot on any laptop is the DC power jack. If someone trips over the power adapter cable while it is still connected to the laptop, there is a high possibility that the power jack will get severely damaged.

On most laptops, the DC power jack is soldered directly to the motherboard and there are only three or four small pins holding it in place, which makes the power jack rather weak. Any sideways pulling of the DC power cord while attached to the laptop will usually dislodge at least one of these pins, breaking the solder around it. Modern laptops use quite a lot of power, from about 70W to 120W or even more. The bad electrical connection from the dislodged pin will cause sparks and heating that will eventually burn a hole through the motherboard and can even be a fire hazard.

Symptoms which indicates poor contact between the DC power jack and the motherboard:

  • The battery is not charging properly or stays at half charge despite having to use the power adaptor.
  • The screen flickers (the brightness is changing) while the power cord is plugged in. This is caused by the laptop switching between DC power (screen is brighter) and battery power (screen is dimmer).
  • The DC plug gets hot after a few minutes of use and may even smell of burning.
  • There are “scratching” sounds coming from the DC jack.

How to test for broken power jack:

    1. Remove the battery

    2. Plug in the power cord

    3. Start the laptop

    4. Gently wiggle the DC power plug on the back of the laptop

If the laptop shuts down abruptly (looses power), find the dealer to re-solder or change the DC power jack as soon as possible as the motherboard has already started to burn around the dislodged pins. The manufacturer might offer to replace the whole motherboard; however the price of a new motherboard together with the labor charge for installing it will usually cost more than the current value of that laptop and sometimes can the pricing can cost over a thousand dollars.

Replacing or re-soldering the DC power jack is not an easy job. It usually takes a few hours depending on the damage and the model of the laptop. To reach the power jack, the laptop has to be completely disassembled and have the motherboard taken out. Then if the board around the dislodged pin is badly burned, the power jack has to be un-soldered and the board has to be patched.

After that a different power jack has to be installed, attached to the case of the laptop and connected with wires to the motherboard, as the patched board wouldn’t be strong enough to support the original type of power jack that was soldered directly to the board.

Hard Disk

Hard disk failures are the most common computer hardware problems. And laptop hard disks fail more often than desktop hard disks due to the portability of a laptop as compared to the rather stagnant laptop. If user knock hard or drop the laptop even a few inches while the hard disk is being accessed or powered up, damage can be done to the hard disk.

Symptoms:

  • The laptop starts as usual but when Windows starts to load it goes to blue screen with either “UNMOUNTABLE_BOOT_VOLUME”, “hal.dll is missing or corrupt” or “WINDOWSSYSTEM32CONFIGSYSTEM” (or similar) is missing or corrupt. All these error messages indicate a file system problem. Sometimes it can be fixed by just scanning the hard disk for errors. However more than half of these cases indicate that the hard disk is starting to go bad and is probably developing bad sectors.
  • The laptop starts as usual but Windows freezes on the initial “Windows XP” screen, although the blue bar continues to move, and users are able to hear feint but continuous clicking. Hard disks have to be replaced.
  • As soon as the laptop is started there are loud clicks or grinding sounds and Windows doesn’t load. Hard disks have to be replaced as well.
  • Shortly after starting the laptop a message appears (usually on a black screen) “SMART test failed. Backup your data immediately and replace the hard disk” or similar. Well, that says it all. S.M.A.R.T. is a self-test that is build into the hard disk.

In almost all cases the laptop can be started from a “live” bootable CD and everything works normal. In rare cases the hard disk PCB (printed circuit board) may get short circuited and even burn. In this scenario, the laptop won’t start unless the hard disk is removed.

A good tool to use to check for hard disk lifespan is ADRC’s hard disk checker. It will scan the hard disk for bad sectors and inform you if the drive is still of good usability.

Overheating

Overheating happens when inadequate airflow from outside the computer’s case across the CPU and motherboard and back out of the case usually via the power supply fan and other case fans.

Once the airflow is disrupted, heat builds up rapidly within the case. This causes components to either shutdown, if such protection is built in or performs erratically. Over the long term, overheating can irreversible damage components.

It is also advisable to keep the computer in a dry and shaded room, which is clean and little dust.

LCD Failures

The most common LCD or “Liquid Crystal Display” screen failure is the inverter, cable and backlight.

Symptoms:

  • The laptop appears to start as usual but the screen is black, if users move it, it turns on and works properly. Failed cable.
  • The laptop appears to start as usual but the screen is very dark, although it changes colour a little and users can see a very feint image. Failed inverter or cable 90%/10%.
  • Immediately after starting the screen has dark pink/red tint and turns dark after a few seconds. Failed backlight.
  • The screen flickers a few times but then turns dark, if users move it or tap it lightly on the sides, it flickers again. Failed cable or LCD 50%/50%.
  • Screen is either black or white (no picture) and external monitor works. Failed cable or LCD 30%/70%.
  • There are stripes on the screen (either vertical or horizontal) that change when users move the screen or tap it lightly on the sides. Failed cable or LCD 10%/90%.

Motherboard

Most motherboard related failures are due to the “On-board” regulated supplies and component failure within those circuits. The on-board power supply circuit had partially failed and was overloading subsequent components else the problem would be with the capacitors which are defective in the first place.

A motherboard failure on a laptop that is out of warranty would usually mean that it’s time for a new laptop. The price of a new motherboard is usually higher than the current value of the laptop.

Symptoms:

Plug in the power adaptor and try to start the laptop.

  • Nothing happens.
  • The “ON” light comes on, the fan spins, but nothing on the screen (black) and you can’t hear the HD working after 10-15 sec.
  • The “ON” light comes on, the fan spins, but after a few seconds the light goes off.
  • After pressing the “ON” button the screen stays black and you can hear a quiet high-pitched sound coming from the laptop.

Corrupted or improper device drivers and operating system

Invalid or corrupt device drivers can wreck havoc on a server, rendering the system unbootable in some cases. Device drivers may also conflict with certain application programs and cause problems with the computer system.

As Windows Vista has just made its launch, chances are some hardware is not supported or not supported fully by the new operating system.

Also, system updates, driver updates can also fix your “hardware” problems.

See updating and fixes

Also, it helps (occasionally) to call the respective support service for help on the driver that is with that particular hardware.

Controller Failure:

A controller failure can act much like a drive failure. However, when a drive fails, users cannot access that particular drive; when a controller fails, users cannot get access to all the drives, devices and any other components that is attached to that particular controller.

A controller fails because a component or components on the on the controller board fails. When users trying to boot the system, they cannot switch on, view, access, or even get hardware conflict messages.

Controller failure occurs due to one of the following reasons:

  • Devices/components not connected properly

      Verify that the cables are properly connecting the device /component to the controller
      If the component fits into a slot, see that the it is flushed in and not loosely connected.
  • Devices/components not setup properly

      Verify that the devices/components is being properly detected and setup in the respective controller setup screen
  • Bad connectors

      Replace the bad connectors with new working ones
  • Bad components

      Replace the component
  • Bad interface board or motherboard

      It is recommended that these components be replaced or that the computer be sent in for servicing

Hardware Failure: The big issue

The big problem, as mentioned earlier, is not replacing or repairing the non-functional hardware, but the time lost, interruptions to daily routine (especially for businesses), and the loss of critical data. If you need a running system at all times, have a backup system as a fail safe, or even two. If users want to ensure that none of their critical data is lost in the event of a hardware problem with their computer, there is only one way to do it — Back up the data as often as possible.

There are many different ways to ensure that the data is backed up, and as technology becomes advances, back up methods become much simpler.

Let’s look for a moment at what a back up is. Backing up of computer files means that users are making a copy of the data that will be stored somewhere other than on the computer. Everyone should always have a back up of at least the most important files, because should anyone experience a hardware problem, they could still be able to access the files.

Many people don’t back up their files, because they don’t know what they need to back up. Start with those files that cannot easily be recreated. Then, move on to files to which you make frequent changes just in case you accidentally change something that you shouldn’t have. Back up your music collection – because this can be quite expensive to replace in the event of data loss.

In general, best practice is to back up the files at least monthly and more frequently if users access the files regularly.