From nobody Thu Mar 14 18:47:01 2024 X-Original-To: freebsd-hardware@mlmmj.nyi.freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2610:1c1:1:606c::19:1]) by mlmmj.nyi.freebsd.org (Postfix) with ESMTP id 4Twbvx6Nf1z5DbLv for ; Thu, 14 Mar 2024 18:47:09 +0000 (UTC) (envelope-from mike@sentex.net) Received: from smarthost1.sentex.ca (smarthost1.sentex.ca [IPv6:2607:f3e0:0:1::12]) (using TLSv1.3 with cipher TLS_AES_256_GCM_SHA384 (256/256 bits) key-exchange X25519 server-signature RSA-PSS (4096 bits) server-digest SHA256 client-signature RSA-PSS (2048 bits) client-digest SHA256) (Client CN "smarthost1.sentex.ca", Issuer "R3" (verified OK)) by mx1.freebsd.org (Postfix) with ESMTPS id 4Twbvw4V8Yz4NGt for ; Thu, 14 Mar 2024 18:47:08 +0000 (UTC) (envelope-from mike@sentex.net) Authentication-Results: mx1.freebsd.org; dkim=none; dmarc=none; spf=pass (mx1.freebsd.org: domain of mike@sentex.net designates 2607:f3e0:0:1::12 as permitted sender) smtp.mailfrom=mike@sentex.net Received: from pyroxene2a.sentex.ca (pyroxene19.sentex.ca [199.212.134.19]) by smarthost1.sentex.ca (8.17.1/8.16.1) with ESMTPS id 42EIl1HG000825 (version=TLSv1.3 cipher=TLS_AES_256_GCM_SHA384 bits=256 verify=FAIL) for ; Thu, 14 Mar 2024 14:47:01 -0400 (EDT) (envelope-from mike@sentex.net) Received: from [IPV6:2607:f3e0:0:4:a1d0:ee73:adac:870f] ([IPv6:2607:f3e0:0:4:a1d0:ee73:adac:870f]) by pyroxene2a.sentex.ca (8.17.1/8.15.2) with ESMTPS id 42EIl0ZH039626 (version=TLSv1.3 cipher=TLS_AES_128_GCM_SHA256 bits=128 verify=NO) for ; Thu, 14 Mar 2024 14:47:00 -0400 (EDT) (envelope-from mike@sentex.net) Message-ID: Date: Thu, 14 Mar 2024 14:47:01 -0400 List-Id: General discussion of FreeBSD hardware List-Archive: https://lists.freebsd.org/archives/freebsd-hardware List-Help: List-Post: List-Subscribe: List-Unsubscribe: Sender: owner-freebsd-hardware@freebsd.org MIME-Version: 1.0 User-Agent: Mozilla Thunderbird Content-Language: en-US To: freebsd-hardware@freebsd.org From: mike tancsa Subject: WD Blue 510 SSD and strange write performance Autocrypt: addr=mike@sentex.net; keydata= xsBNBFywzOMBCACoNFpwi5MeyEREiCeHtbm6pZJI/HnO+wXdCAWtZkS49weOoVyUj5BEXRZP xflV2ib2hflX4nXqhenaNiia4iaZ9ft3I1ebd7GEbGnsWCvAnob5MvDZyStDAuRxPJK1ya/s +6rOvr+eQiXYNVvfBhrCfrtR/esSkitBGxhUkBjOti8QwzD71JVF5YaOjBAs7jZUKyLGj0kW yDg4jUndudWU7G2yc9GwpHJ9aRSUN8e/mWdIogK0v+QBHfv/dsI6zVB7YuxCC9Fx8WPwfhDH VZC4kdYCQWKXrm7yb4TiVdBh5kgvlO9q3js1yYdfR1x8mjK2bH2RSv4bV3zkNmsDCIxjABEB AAHNHW1pa2UgdGFuY3NhIDxtaWtlQHNlbnRleC5uZXQ+wsCOBBMBCAA4FiEEmuvCXT0aY6hs 4SbWeVOEFl5WrMgFAl+pQfkCGwMFCwkIBwIGFQoJCAsCBBYCAwECHgECF4AACgkQeVOEFl5W rMiN6ggAk3H5vk8QnbvGbb4sinxZt/wDetgk0AOR9NRmtTnPaW+sIJEfGBOz47Xih+f7uWJS j+uvc9Ewn2Z7n8z3ZHJlLAByLVLtcNXGoRIGJ27tevfOaNqgJHBPbFOcXCBBFTx4MYMM4iAZ cDT5vsBTSaM36JZFtHZBKkuFEItbA/N8ZQSHKdTYMIA7A3OCLGbJBqloQ8SlW4MkTzKX4u7R yefAYQ0h20x9IqC5Ju8IsYRFacVZconT16KS81IBceO42vXTN0VexbVF2rZIx3v/NT75r6Vw 0FlXVB1lXOHKydRA2NeleS4NEG2vWqy/9Boj0itMfNDlOhkrA/0DcCurMpnpbM7ATQRcsMzk AQgA1Dpo/xWS66MaOJLwA28sKNMwkEk1Yjs+okOXDOu1F+0qvgE8sVmrOOPvvWr4axtKRSG1 t2QUiZ/ZkW/x/+t0nrM39EANV1VncuQZ1ceIiwTJFqGZQ8kb0+BNkwuNVFHRgXm1qzAJweEt RdsCMohB+H7BL5LGCVG5JaU0lqFU9pFP40HxEbyzxjsZgSE8LwkI6wcu0BLv6K6cLm0EiHPO l5G8kgRi38PS7/6s3R8QDsEtbGsYy6O82k3zSLIjuDBwA9GRaeigGppTxzAHVjf5o9KKu4O7 gC2KKVHPegbXS+GK7DU0fjzX57H5bZ6komE5eY4p3oWT/CwVPSGfPs8jOwARAQABwsB2BBgB CAAgFiEEmuvCXT0aY6hs4SbWeVOEFl5WrMgFAl+pQfkCGwwACgkQeVOEFl5WrMiVqwf9GwU8 c6cylknZX8QwlsVudTC8xr/L17JA84wf03k3d4wxP7bqy5AYy7jboZMbgWXngAE/HPQU95NM aukysSnknzoIpC96XZJ0okLBXVS6Y0ylZQ+HrbIhMpuQPoDweoF5F9wKrsHRoDaUK1VR706X rwm4HUzh7Jk+auuMYfuCh0FVlFBEuiJWMLhg/5WCmcRfiuB6F59ZcUQrwLEZeNhF2XJV4KwB Tlg7HCWO/sy1foE5noaMyACjAtAQE9p5kGYaj+DuRhPdWUTsHNuqrhikzIZd2rrcMid+ktb0 NvtvswzMO059z1YGMtGSqQ4srCArju+XHIdTFdiIYbd7+jeehg== Content-Type: text/plain; charset=UTF-8; format=flowed Content-Transfer-Encoding: 8bit X-Scanned-By: MIMEDefang 2.86 on 64.7.153.18 X-Spamd-Bar: --- X-Spamd-Result: default: False [-3.37 / 15.00]; NEURAL_HAM_LONG(-1.00)[-1.000]; NEURAL_HAM_MEDIUM(-1.00)[-1.000]; NEURAL_HAM_SHORT(-0.98)[-0.985]; R_SPF_ALLOW(-0.20)[+ip6:2607:f3e0::/32]; MIME_GOOD(-0.10)[text/plain]; RCVD_IN_DNSWL_LOW(-0.10)[199.212.134.19:received]; XM_UA_NO_VERSION(0.01)[]; RCPT_COUNT_ONE(0.00)[1]; ASN(0.00)[asn:11647, ipnet:2607:f3e0::/32, country:CA]; MID_RHS_MATCH_FROM(0.00)[]; MIME_TRACE(0.00)[0:+]; FREEFALL_USER(0.00)[mike]; R_DKIM_NA(0.00)[]; MLMMJ_DEST(0.00)[freebsd-hardware@freebsd.org]; DMARC_NA(0.00)[sentex.net]; FROM_EQ_ENVFROM(0.00)[]; FROM_HAS_DN(0.00)[]; ARC_NA(0.00)[]; RCVD_COUNT_TWO(0.00)[2]; TO_MATCH_ENVRCPT_ALL(0.00)[]; TO_DN_NONE(0.00)[]; PREVIOUSLY_DELIVERED(0.00)[freebsd-hardware@freebsd.org]; RCVD_TLS_ALL(0.00)[] X-Rspamd-Queue-Id: 4Twbvw4V8Yz4NGt This might be more of a hardware question than anything, but I noticed the drive is fairly fast on initial writes, but dramatically slows down over time with a consistent write. At bootup time, I can blast out a file (UFS2 mount) #  dd if=/dev/zero of=/mnt/tmp/junk.bin.`date "+%s"` bs=1m status=progress count=4000   3735027712 bytes (3735 MB, 3562 MiB) transferred 7.014s, 533 MB/s # nice and fast.  But subsequent writes really start to tank. #  dd if=/dev/zero of=/mnt/tmp/junk.bin.`date "+%s"` bs=1m status=progress count=4000   4128243712 bytes (4128 MB, 3937 MiB) transferred 46.048s, 90 MB/s # dd if=/dev/zero of=/mnt/tmp/junk.bin.`date "+%s"` bs=1m status=progress count=4000   3992977408 bytes (3993 MB, 3808 MiB) transferred 31.016s, 129 MB/s If I wait for 2min, it seems to be back to normal  # sleep 120 ; dd if=/dev/zero of=/mnt/tmp/junk.bin.`date "+%s"` bs=1m status=progress count=4000   4137680896 bytes (4138 MB, 3946 MiB) transferred 9.025s, 458 MB/s Is there something going on behind the scenes limiting the amount of sustained writes these drives can handle ? Is it just a limitation of the SSD ?  I didnt notice such issues on some consumer Samsungs. The problem initially showed up for me when I was doing a series of zfs send | zfs recv on the same pool (so a lot of reads and writes at the same time) and I would get a bunch of errors on the WD disks.  Replacing them with Samsungs avoids the errors. === START OF INFORMATION SECTION === Device Model:     WD Blue SA510 2.5 1000GB Serial Number:    240406800001 LU WWN Device Id: 5 001b44 8b334e00b Firmware Version: 52046100 User Capacity:    1,000,204,886,016 bytes [1.00 TB] Sector Size:      512 bytes logical/physical Rotation Rate:    Solid State Device Form Factor:      2.5 inches TRIM Command:     Available, deterministic Device is:        Not in smartctl database 7.3/5528 ATA Version is:   ACS-4, ACS-2 T13/2015-D revision 3 SATA Version is:  SATA 3.2, 6.0 Gb/s (current: 6.0 Gb/s) Local Time is:    Thu Mar 14 14:41:58 2024 EDT SMART support is: Available - device has SMART capability. SMART support is: Enabled # camcontrol iden ada1 pass1: ACS-4 ATA SATA 3.x device pass1: 600.000MB/s transfers (SATA 3.x, UDMA6, PIO 512bytes) protocol              ACS-4 ATA SATA 3.x device model          WD Blue SA510 2.5 1000GB firmware revision     52046100 serial number         240406800001 WWN                   5001b448b334e00b additional product id cylinders             16383 heads                 16 sectors/track         63 sector size           logical 512, physical 512, offset 0 LBA supported         268435455 sectors LBA48 supported       1953525168 sectors PIO supported         PIO4 DMA supported         WDMA2 UDMA6 media RPM             non-rotating Zoned-Device Commands no Feature                      Support  Enabled   Value Vendor read ahead                     yes      yes write cache                    yes      yes flush cache                    yes      yes Native Command Queuing (NCQ)   yes              32 tags NCQ Priority Information       no NCQ Non-Data Command           no NCQ Streaming                  no Receive & Send FPDMA Queued    no NCQ Autosense                  no SMART                          yes      yes security                       yes      no power management               yes      yes microcode download             yes      yes advanced power management      yes      no      0/0x00 automatic acoustic management  no       no media status notification      no       no power-up in Standby            no       no write-read-verify              no       no unload                         no       no general purpose logging        yes      yes free-fall                      no       no sense data reporting           no       no extended power conditions      no       no device statistics notification no       no Data Set Management (DSM/TRIM) yes DSM - max 512byte blocks       yes              8 DSM - deterministic read       yes              any value Trusted Computing              no encrypts all user data         no Sanitize                       yes              block, Sanitize - commands allowed    yes Sanitize - antifreeze lock     yes Host Protected Area (HPA)      no Accessible Max Address Config  yes      no 1953525168/1953525168 RELENG_14 from today