Gazduirea pe baza de dispozitive ARM a devenit o realitate (in cele din urma)

De cativa ani tot citesc in publicatiile de specialitate posibilitatea de a folosi servere ARM, despre o tehnologie ce urmeaza sa acapareze zeci de procente din piata de gazduire, despre potentialul incredibil al microserverelor sau despre cantitatea de curent electric infima pe care astfel de servere o consuma.

Insa…. realitatea este ca acest gen de serviciu(gazduirea pe servere ARM) s-a cam lasat asteptat, iar la sfarsitul anului 2015 nu am gasit decat un singur provider ce si-a orientat business-ul catre acest gen de tehnologie. Daca merita sau nu… veti putea vedea in benchmarkul de mai jos(cu toate ca rezultatele nu sunt catastrofale, iar faptul ca mai jos de atat nu se poate merge – memoria si procesorul fiind dedicate):

========================================================================
   BYTE UNIX Benchmarks (Version 5.1.3)

   System: ******************
   OS: GNU/Linux -- 3.2.34-29 -- #16 SMP Thu Mar 12 09:45:43 UTC 2015
   Machine: armv7l (armv7l)
   Language: en_US.utf8 (charmap="UTF-8", collate="UTF-8")
   11:00:33 up 53 min,  2 users,  load average: 0.08, 0.05, 0.09; runlevel 2

------------------------------------------------------------------------
Benchmark Run: Wed Apr 15 2015 11:00:33 - 11:28:50
0 CPUs in system; running 1 parallel copy of tests

Dhrystone 2 using register variables        3659559.7 lps   (10.0 s, 7 samples)
Double-Precision Whetstone                      558.5 MWIPS (10.0 s, 7 samples)
Execl Throughput                                939.9 lps   (29.9 s, 2 samples)
File Copy 1024 bufsize 2000 maxblocks        110662.5 KBps  (30.0 s, 2 samples)
File Copy 256 bufsize 500 maxblocks           31229.5 KBps  (30.0 s, 2 samples)
File Copy 4096 bufsize 8000 maxblocks        291155.9 KBps  (30.0 s, 2 samples)
Pipe Throughput                              196414.4 lps   (10.0 s, 7 samples)
Pipe-based Context Switching                  34778.2 lps   (10.0 s, 7 samples)
Process Creation                               2721.9 lps   (30.0 s, 2 samples)
Shell Scripts (1 concurrent)                   2011.4 lpm   (60.0 s, 2 samples)
Shell Scripts (8 concurrent)                    610.5 lpm   (60.1 s, 2 samples)
System Call Overhead                         677934.5 lps   (10.0 s, 7 samples)

System Benchmarks Index Values               BASELINE       RESULT    INDEX
Dhrystone 2 using register variables         116700.0    3659559.7    313.6
Double-Precision Whetstone                       55.0        558.5    101.5
Execl Throughput                                 43.0        939.9    218.6
File Copy 1024 bufsize 2000 maxblocks          3960.0     110662.5    279.5
File Copy 256 bufsize 500 maxblocks            1655.0      31229.5    188.7
File Copy 4096 bufsize 8000 maxblocks          5800.0     291155.9    502.0
Pipe Throughput                               12440.0     196414.4    157.9
Pipe-based Context Switching                   4000.0      34778.2     86.9
Process Creation                                126.0       2721.9    216.0
Shell Scripts (1 concurrent)                     42.4       2011.4    474.4
Shell Scripts (8 concurrent)                      6.0        610.5   1017.4
System Call Overhead                          15000.0     677934.5    452.0
                                                                   ========
System Benchmarks Index Score                                         264.9

------------------------------------------------------------------------
Benchmark Run: Wed Apr 15 2015 11:28:50 - 11:57:11
0 CPUs in system; running 4 parallel copies of tests

Dhrystone 2 using register variables       14622411.6 lps   (10.0 s, 7 samples)
Double-Precision Whetstone                     2234.1 MWIPS (10.0 s, 7 samples)
Execl Throughput                               3123.9 lps   (29.9 s, 2 samples)
File Copy 1024 bufsize 2000 maxblocks        176578.5 KBps  (30.0 s, 2 samples)
File Copy 256 bufsize 500 maxblocks           49194.3 KBps  (30.0 s, 2 samples)
File Copy 4096 bufsize 8000 maxblocks        523858.4 KBps  (30.0 s, 2 samples)
Pipe Throughput                              668236.2 lps   (10.0 s, 7 samples)
Pipe-based Context Switching                 187468.2 lps   (10.0 s, 7 samples)
Process Creation                               8417.6 lps   (30.0 s, 2 samples)
Shell Scripts (1 concurrent)                   4988.5 lpm   (60.0 s, 2 samples)
Shell Scripts (8 concurrent)                    642.3 lpm   (60.3 s, 2 samples)
System Call Overhead                        2420990.4 lps   (10.0 s, 7 samples)

System Benchmarks Index Values               BASELINE       RESULT    INDEX
Dhrystone 2 using register variables         116700.0   14622411.6   1253.0
Double-Precision Whetstone                       55.0       2234.1    406.2
Execl Throughput                                 43.0       3123.9    726.5
File Copy 1024 bufsize 2000 maxblocks          3960.0     176578.5    445.9
File Copy 256 bufsize 500 maxblocks            1655.0      49194.3    297.2
File Copy 4096 bufsize 8000 maxblocks          5800.0     523858.4    903.2
Pipe Throughput                               12440.0     668236.2    537.2
Pipe-based Context Switching                   4000.0     187468.2    468.7
Process Creation                                126.0       8417.6    668.1
Shell Scripts (1 concurrent)                     42.4       4988.5   1176.5
Shell Scripts (8 concurrent)                      6.0        642.3   1070.6
System Call Overhead                          15000.0    2420990.4   1614.0
                                                                   ========
System Benchmarks Index Score                                         705.2

Despre… ghost writer

Nu am sa stau sa va povestesc despre mine, despre cine sunt si despre hobbie-urile mele. Omul din spatele Ghost Writer nu este important – de fapt de aici vine si ideea de ghost, nu-i asa? – dar daca doriti sa ma deconspirati, sunt convins ca nu este nevoie de un detectiv particular pentru a afla cine se afla in spatele acestor randuri.

Proiectul de fata nu o sa fie un blog in care sa postez despre pisica vecinilor, invazia sirienilor, politica, moda si nici despre alte subiecte de larg interes – scopul proiectului este de a scoate in fata informatii din “industria online” – fie ele de natura didactica, tehnica, fie diverse idei ce isi vor gasi un refugiu in paginile acestui blog.

Ca si plan pentru ceea ce a mai ramas din anul 2015, as vrea sa abordez cateva subiecte pentru incepatori, subiecte usoare, prin care voi incerca sa-mi aduc aminte cum se scrie un text si cum ar trebui exprimate in scris cateva dintre notiunile pe care ar trebui sa le cunoasca orice persoana ce doreste o cariera in mediul online.

Din cuprins:

Capitolul 1 – “In cautarea bunastarii”.
– De la idee la realizarea unui proiect: cum se face studiul de “fezablitate”?
– Metode de evaluare a potentialului online (al unei afaceri)
– Efectuarea unui keyword research complet si corect…
– Este promovarea prin socialmedia utila? Daca da, cand? Si pentru cine?
– Cum sa facem spam? Adica… e-mail marketing.