Qual é a maneira mais eficiente de fazer backup dos sistemas de uma grande organização?

1st of all you've to decide what type of data you need to backup, and the size of data, and how important is it. then what type of backup method you could go with. the last not least you is it local backup or on the cloud.

List the systems with corresponding category then design the desired backup plan and retention. Select the backup system based on the available capacity , cost efficiency & criticality position . Simulate the backup procedure every now and then to maintain the backup setup whenever needed.

If you look at the world and the way forward, data is at the centre of everything. It will shape business, humans, technology, and decision-making in the future. I suggest the following: 1. Invest in a proven data backup system and even consider a dedicated engineer responsible for Data backup, restoration, and integrity of the process 2. Offsite storage of data backup. 3. DRP for the organisation which should tie into the organisation's BCP. A lot of organisations gave less attention to business risk, and as a result, were caught off guard. A lot of business in the pandemic, natural or social unrest cannot continue and work from home without an investment in DRP.

Hay que seguir el esquema tres dos uno. Tres copias, en dos soportes, y una de las fuera de la organización. De esta manera se minimizan las posibilidades de no disponer de un buen conjunto de copias para su restauración.

I'd say since the majority of organisations don't classify or categorise their data..... . and most likely no designated information system ownership......and most likely don't even know what data they have; EVERYTHING Just back up every single byte of data you can find. Stretch that elastic storage like an obese person who just ate at that Monty Python restaurant's belt. You know, till it's so taut, its 𝕎𝕒𝕗𝕖𝕣 𝕥𝕙𝕚𝕟 After that, I reckon tapes are pretty good if you're physically able. Scotch is a good one. I wouldn't worry about RTO/RPO, cause no one usually tests the backups. Restoration is futile. Good luck! 🤖

Configure a weekly schedule for comprehensive data capture through full backups. Followed by a daily incremental backups to capture changes since the last backup. This combination ensures a thorough and efficient backup strategy.

Customers should select a solution that provides the next-generation backup methodologies- a tool that creates a full backup copies and continues the subsequent backups via the incremental forever option that uses the B-Tree referencing option in storing the backup sets. Most of the newer backup providers in the market leverage this option. Additionally, selecting a native-saas solution would remove the management overhead that typically comes with running and supporting a backup infrastructure. Backup copies should be stored in an offside location that is entirely air-gapped and immutable. There should also be an option to create a replica copy of the backup that could be moved to a lower storage tier as an archive copy.

An entrepreneurial ransomware gang may be prepared to strike a recurring deal for offsite backups with better customer support and lower rates than many big-box vendors.

Various backup methods are in operation in the current context. It's essential you choose the right backup methodology to ensure the data safety is guaranteed, and the day-to-day work isn't interrupted.

Choisir la bonne méthode de sauvegarde dépend des besoins de l'organisation. Les sauvegardes complètes, bien que fiables, sont coûteuses en temps et en stockage. Les sauvegardes incrémentielles, plus rapides, enregistrent uniquement les modifications depuis la dernière sauvegarde, optimisant le stockage et le temps, mais compliquent la restauration. Les sauvegardes différentielles, un compromis, sauvegardent les changements depuis la dernière sauvegarde complète. Elles offrent un équilibre entre espace de stockage, vitesse et simplicité de restauration. Évaluez la fréquence des modifications de données pour choisir la méthode adaptée.

Automatizar e agendar backups é crucial porque garante a regularidade e consistência na proteção dos dados. A automação reduz o risco de esquecimento humano, enquanto o agendamento permite a realização periódica sem intervenção manual, assegurando a integridade das informações em caso de falhas ou perda de dados.

Automatizar é a palavra que move as áreas de tecnologia. Automatizando se garante que seja feito, que seja monitorado e que o time possa utilizar seu tempo em pesquisa e busca de melhorias.

No puedes caer en el error de programar tus copias y olvidarte. Diariamente debes chequear que se realizan correctamente. Además debes tener un calendario para planificar restauraciones para verificar el proceso completo. He visto muchos casos en los que han intentado restaurar copias de seguridad y se han llevado un gran disgusto al ver que no están disponibles... También es muy importante contar con un plan de recuperación bien definido. Hay que tener claro donde restaurar, si tenemos espacio para hacerlo sin machacar los datos existentes, como restaurar las copias si el programa con el las restauras está dañado,...

I want to put it short. Most efficient way is to write your own script via bash/powershell and automate everything. Via scripting you avoid huge process taking commercial solutions plus saving your budget.

The biggest benefit to automating any schedule is that it reduces the off chance of human error. It also means that more components can be included without necessarily incurring extra overhead. For example, combining data management capabilities like instant, point-in-time snapshots can bolster a more robust weekly or monthly full backup without impacting recoverability or performance.

Using Tiered Storage and Lifecycle Management. Actually, Tiered storage reduces costs by utilizing high-performance storage for critical data and more cost-effective storage for less critical or archival data. On the other side Lifecycle policy ensures that you retain only necessary data, reducing storage costs.

This is actually one area where public cloud is incredibly valuable. Backup tends to be less frequently accessed than primary data (often it's only when you're restoring) and so it makes sense to offload that data either directly into the cloud, or tiered off so that older backups can be retained for a longer period on colder storage. Optimising storage also means reviewing what has been backed up on a regular basis to understand if it's still required. Deleting backups can be scary, but it can also prevent potential regulation breaches (for example, data that is only supposed to be held for a certain amount of time).

Optimize and save money, those key points are more and more becoming requirements for organizations from small to big enterprises.

Con el tiempo, el coste del almacenamiento se ha reducido bastante por lo que no debe ser obstáculo. Pero sigue siendo interesante reducirlo sobre todo para copias en la nube, ya que no solamente tendremos coste del almacenamiento, si no que la velocidad no es tan alta como en local. Respecto al cifrado lo considero imprescindible si quieres dotar tus copias de un mínimo de seguridad

I don't believe in deduplication for critical data. As storage space becomes a commodity, I have no problem using more of it rather than risk the integrity (the big I of the CIA Triade) of my data. Deduplication means that if data (e.g. files or blocks) is found multiple times, only a single copy is kept. All other copies are deleted and replaced by a pointer to the single copy. Deduplication artificially increases the data density and therefore the criticality of each bit stored on the hard drive. If a tiny bit of data is corrupted (e.g. by a power failure, an uncorrected bit flip or a firmware bug), deduplication would spread it like wildfire. If the only copy is invalid, all pointers to it become stale.

To back up a large organization's systems in the best way, you need to do the following things: make regular incremental backups, use automated tools, store your backups in the cloud for more space and safety, and separate your data into different levels of importance for backup purposes.

The cloud-based backup approach is very effective because it transfers the majority of risk away from the organization and provides easy access to backed-up data from anywhere. However, we should be careful to always do traditional local backups of the cloud environment at intervals (monthly or quarterly) to fully close any potential data loss.

Cloud storage is cheap, but factor in costs for bandwidth, ingres and egress times particularly with multiple backups running concurrently should also be part of evaluation process. Cloud storage whilst it has short term cost savings of can often exceed on premise solutions. Consider the total cost of ownership.

A very popular topic these days. Part of leading practice for backup solutions is offsite backups and immutability to safeguard against ransomware. Using Object Storage is a cheaper and secure way to store your backups. Mot of the current popular backup solutions offers a way to replicate your backup jobs to an object storage bucket on public or private cloud. In most cases copy the URL of the bucket into the backup software and voila. Both Azure and AWS offer container lock on their object storage. Other vendors like Wasabi also offer immutability options.

Data backup & restoration is a crucial process for any large organization, especially if it operates in a regulated industry. Depending on the type of data and the industry, there may be different rules and standards to follow for data backup. For example, financial, banking, or payment card organizations have to comply with PCI DSS standards, while pharmaceutical or healthcare organizations have to follow CFR 21 or HIPAA regulations. Moreover, data backup has to take into account the country-specific regulations and laws that apply to the organization. Data backup is not only about protecting the data from loss or damage, but also about restoring it quickly and effectively in case of a disaster after following specific regulations & laws.

Un tema muy importante relativo a los backups es que se han de realizar pruebas fiables. Con pruebas fiables nos referimos a aquellas que simulan lo que puede ser necesario hacer en un entorno de desastre, no simplemente recuperar cursos archivos o un sistema aislado. Tan importante es ser capaz de recuperar la información y tener un claro procedimiento para realizarlo, en base a los sierra escenarios que se puedan presentar como tener una política adecuada de backup.

Backups are only as good as the restoration process vision. If you define what you expect for data restore that will drive which way you plan your backups.

Backup is only useful if tested regularly for all critical services Don't forget : You don't need a backup plan You need a Restore plan

To efficiently back up a large organization's systems data involves a mix of automated and incremental backups. Choose a idempotent solution such as cloud-based storage, implement strong encryption for data security, and have an offsite storage for disaster recovery. Regularly test and validate backups, monitor success and maintain up-to-date documentation. Prioritize efficiency by keeping in mind the strategy to the organization's unique needs, and periodically review and update procedures from an agile perspective for optimal data protection and system resilience.

There is also a need to be predictive and monitor the growth of the DB. One needs to be able to plan ahead to avoid supprise incase the backups are local hence a cloud base solution is always recommend depending on the data privacy laws of the country