[Git][venenux/venenux-docus][master] the vertica lie vs RDBMS - real true and use cases must be for BI

0 views
Skip to first unread message

Герхард PICCORO Lenz McKAY (@mckaygerhard)

unread,
Jun 1, 2023, 1:07:38 AM6/1/23
to venenux...@googlegroups.com

Герхард PICCORO Lenz McKAY pushed to branch master at venenux / venenux-docus

Commits:

  • 94dd86c0
    by Герхард PICCORO Lenz McKAY at 2023-06-01T05:07:32+00:00
    the vertica lie vs RDBMS - real true and use cases must be for BI

1 changed file:

Changes:

  • docs/others-infodocs-details-vertica-vs-opendbms-en.md
    1
    +# Vertica vs RDBMS and noSQL database cases
    
    2
    +
    
    3
    +Document/key-value databases is typically good for unstructured/"schemaless" 
    
    4
    +data. Usually, the cases when you don't need to explicitly define your schema 
    
    5
    +up front and can just include new fields without any ceremony. Most modern 
    
    6
    +fiscal printers uses such way of storage in the client machine.
    
    7
    +
    
    8
    +In recents years those kind of databases where a boom, previously we have been 
    
    9
    +on the benealt of LevelDB but such engine was very unstable when crashing.
    
    10
    +
    
    11
    +It's often very easy to scale out document/key-value databases. Just by more 
    
    12
    +parts of same strutures (knows as nodes) to replicate data to is one way to 
    
    13
    +offer more scalability and offer more protection against data loss.
    
    14
    +
    
    15
    +Otherwise, complex/dynamic queries/reporting are best served from an RDBMS. Often 
    
    16
    +
    
    17
    +#### Introduction to Vertica DB
    
    18
    +
    
    19
    +**Vertica** breaks the database scenario for two things that are the first things 
    
    20
    +technology managers really see, **its level of data compression (both in data 
    
    21
    +storage as well as in transmission) as well as the semi-free service** what 
    
    22
    +are offering. But all of those are great only if the company that implements 
    
    23
    +their technology is already also a company of technology. 
    
    24
    +
    
    25
    +But what if the company trying to make new high data writing software with high 
    
    26
    +volume data storage isn't primarily focused on technology and can't spend that 
    
    27
    +much on such a thing? But part of the information os not so true at all, is 
    
    28
    +not open source per se. Also we have to take into account today's world of 
    
    29
    +compulsive political tension!
    
    30
    +
    
    31
    +#### DBMS like percona mysql with xtradb and myrocks
    
    32
    +
    
    33
    +You may think of Percona as a distributor who collects, coordinates and maintains 
    
    34
    +patches and distributes an enhanced version of the MySQL server.
    
    35
    +
    
    36
    +Inspired in RocksDB, **myrocks** is a percona improved engine that permits 
    
    37
    +to use SQL like request and client into the MySQL DBMS.
    
    38
    +With SSD database storage, this means less space used and a higher endurance 
    
    39
    +of the storage over time, but lack of performance if the underliying 
    
    40
    +storage hardware is not SSD, so RocksDB need several key OS+HArdware 
    
    41
    +features to have results.
    
    42
    +
    
    43
    +The library is maintained by the Facebook Database Engineering Team.
    
    44
    +It is a fork of Google's LevelDB optimized to exploit many CPU cores, 
    
    45
    +and make efficient use of fast storage, such as solid-state drives (SSD), 
    
    46
    +for input/output (I/O) bound workloads.
    
    47
    +
    
    48
    +The **XtraDB** is InnoDB with steroids. The patches themselves stem from 
    
    49
    +Google, Facebook and others. XtraDB scales better on massively parallel architectures 
    
    50
    +and especially XtraDB is much better suited for write-heavy workload. Contrary 
    
    51
    +to rocksDB its well suite to use in HDD or SDD, but in last will cut off the 
    
    52
    +SDD storage life due workload. XtraDB includes features such as crash-safe 
    
    53
    +replication, online backups, and hot backups that are not available in InnoDB. 
    
    54
    +
    
    55
    +However, it's also important to note that XtraDB is not compatible with some 
    
    56
    +InnoDB features, such as full-text searching or spatial indexing.
    
    57
    +
    
    58
    +## Considerations
    
    59
    +
    
    60
    +#### Considerations for Percona flavours
    
    61
    +
    
    62
    +Percona XtraDB is compatible with InnoDB by default. You can read and write the 
    
    63
    +same datafiles, and all SQL queries run exactly the same. You won’t even notice 
    
    64
    +the difference.
    
    65
    +
    
    66
    +Percona PosggreSQL does not have many differences, just better integration and 
    
    67
    +easy of usage for administrators. This cos PostgreSQL is so complex and so scalar 
    
    68
    +already.
    
    69
    +
    
    70
    +The improvements in Percona Mysql XtraDB are subtle. They are internal fixes to 
    
    71
    +solve specific scaling bottlenecks. These bottlenecks don’t necessarily affect 
    
    72
    +your applications or environment, in which case Percona XtraDB would function 
    
    73
    +exactly like stock InnoDB. Some of the enhancements in Percona XtraDB proved 
    
    74
    +themselves useful, so later versions of Oracle MySQL and MariaDB implemented 
    
    75
    +and today XtraDB and InnoDB are ver similar.
    
    76
    +
    
    77
    +* Buffer pool mutex split into four sub-types of mutex, to reduce contention when 
    
    78
    +you have a high number of concurrent clients.
    
    79
    +* Insert buffer options for max size and merge rate. Good when you have lots of 
    
    80
    +indexes and a very high rate of insert/update/delete operations.
    
    81
    +* Adaptive hash may be split into multiple partitions. Good if you have a high 
    
    82
    +number of threads running concurrent queries over non-primary indexes, so much 
    
    83
    +that it’s causing contention on the Adaptive Hash Index mutex.
    
    84
    +* Faster page checksum algorithm. Good if you have a high rate of page flushes 
    
    85
    +on SSD storage. This feature is obsolete in MySQL 5.6.
    
    86
    +* Handle corrupt tables by issuing a warning and marking the table unusable, 
    
    87
    +instead of the default behavior of deliberately crashing the MySQL server.
    
    88
    +
    
    89
    +#### considerations for Vertica DB
    
    90
    +
    
    91
    +The biggest advantage of Vertica is the raw speed. It is extremely fast when 
    
    92
    +compared to other analytical databases, and it boasts features that make joins 
    
    93
    +extremely fast. The trade-offs are the high license costs, slow updates and 
    
    94
    +insatiable need to eat thru tons of disk space.
    
    95
    +
    
    96
    +Vertica never overwrite the data file on updates, so every time you update and 
    
    97
    +new SO write will happen. This is a cons of the key-value storage philosophy, 
    
    98
    +and for a relational base data this is not the good option.
    
    99
    +
    
    100
    +Vertica is not to be used to replace your relational data at OLTP database, this 
    
    101
    +is there to do the heavy lifting and help you do the analytics stuff with less 
    
    102
    +expense(time, money). Vertica its more for BI rather for information storage.
    
    103
    +
    
    104
    +Vertica have compressed data and encoded data, the compressed data will require 
    
    105
    +some extra cpu cycles while retrieved, but most of the times Vertica uses encoding 
    
    106
    +as following source describes http://www.aodba.com/tut_output_mysql.php?tut=6&page=vertica 
    
    107
    +encoding creates a smaller footprints and by doing this data retrieval will be faster.
    
    108
    +
    
    109
    +For a storage inventary software or product book is not the right choose.
    
    110
    +Vertica is a BI tool basically.
    
    111
    +
    
    112
    +Vertica is extremely expensive, it not opensource, the opensource version does 
    
    113
    +not have all of this key features; prices it started with about 100K$ per terabyte, 
    
    114
    +and after it became popular now it costs 10K$ per terabyte, still this is expensive, 
    
    115
    +a fully protected server is worth less than 500$ per month including several 
    
    116
    +terabytes in Germany.
    
    117
    +
    
    118
    +#### the lie of Vertica open source
    
    119
    +
    
    120
    +Community version will only allow you to create a 3 nodes cluster and a max 
    
    121
    +of 1TB (no bkp available and others things are as well not possible)
    
    122
    +
    
    123
    +Just to put some light into the Vertica licenses
    
    124
    +
    
    125
    +- All development,homologation or desaster sites (they are under the initial 
    
    126
    +production License – no extra money)
    
    127
    +- Raw data size based (valid to store up to some amount of raw data), license is 
    
    128
    +only applied to the RAW data loaded once loaded it can be replicated as many times 
    
    129
    +as you like.
    
    130
    +- Term-based and peridicaly (valid until a specific date, montly by example).
    
    131
    +
    
    132
    +
    
    133
    +## Basic Table comparison
    
    134
    +
    
    135
    +| Topics       | Vertica                            | Percona Mysql+Myrocks               |
    
    136
    +| ------------ | ---------------------------------- | ----------------------------------- |
    
    137
    +| TODO         | TODO                               | TODO                                |
    
    138
    +
    
    139
    +
    
    140
    +## Conclusions
    
    141
    +
    
    142
    +SQL performs well as a transaction processing system, it works horrible when 
    
    143
    +trying to query it for reporting/analytic purposes. Tha is the case for Vertica, 
    
    144
    +and all the colum store capable (cos Vertica is primary RDBMS, not column/key based) 
    
    145
    +has advantages but not as sustitution for.
    
    146
    +
    
    147
    +You should choose Vertica if you need reporting and consulting features, mostly 
    
    148
    +those need at bussines inteligence, but not like invetary or storage time.
    
    149
    +
    
    150
    +This means any column/key based DB is just complementary to, not sustitute to, and 
    
    151
    +is the reason why Vertica really still is a RDBMS and not columnkey only.
    
    152
    +


View it on GitLab.
You're receiving this email because of your account on gitlab.com. Manage all notifications · Help

Reply all
Reply to author
Forward
0 new messages