pulsar-ack-1| 3.23 | 3.21 | 4.23 | 4.73 | 5.89 | pulsar-ack-2| 3.64 | 3.42 | 4.47 | 10.37 | 32.74 | kafka-ack-1| 2.54 | 1.99 | 3.78 | 11.37 | 18.71 | kafka-ack-all| 9.84 | 5.71 | 19.7 | 58.83 | 164.20"}}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"爲了更深入地瞭解延遲如何隨時間變化，我們繪製了 Pulsar 和 Kafka 採取不同複製持久性設置時的 P99 延遲圖。如圖 11 所示，Pulsar 的延遲比較穩定（約 5 毫秒），但 Kafka 的延遲波動較大。對於關鍵任務服務（mission-critical services）來說，穩定、持續的低延遲至關重要。"}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"image","attrs":{"src":"https:\/\/static001.infoq.cn\/resource\/image\/9f\/5b\/9f49d0eabfeb36910b49b74a75d8f85b.png","alt":null,"title":null,"style":[{"key":"width","value":"75%"},{"key":"bordertype","value":"none"}],"href":null,"fromPaste":true,"pastePass":true}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":"center","origin":null},"content":[{"type":"text","text":"圖11 Pulsar 和 Kafka 的 P99 延遲"}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","marks":[{"type":"strong"}],"text":"端到端延遲--同步本地持久性"}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"使用兩種複製持久性設置（分別爲 ack-1和 ack-2）和同步本地持久性時，Pulsar 和 Kafka 端到端延遲的差異如圖 12。表 3 爲每種情況下的精確延遲數值。在異步複製持久性（ack-1）下，Pulsar P99 端到端延遲是 Kafka 的 1\/3，在同步複製持久性（ack-2）下是 Kafka 的 1\/5。"}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"image","attrs":{"src":"https:\/\/static001.infoq.cn\/resource\/image\/7d\/3a\/7d46b026cc1d195e535a47d5f1f42e3a.png","alt":null,"title":null,"style":[{"key":"width","value":"75%"},{"key":"bordertype","value":"none"}],"href":null,"fromPaste":true,"pastePass":true}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":"center","origin":null},"content":[{"type":"text","text":"圖12 1 個訂閱時，Pulsar 和 Kafka 的端到端延遲（數據同步）"}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","marks":[{"type":"strong"}],"text":"表 3 1 個訂閱時，Pulsar 和 Kafka 的實際端到端延遲測試結果(數據同步)"}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"embedcomp","attrs":{"type":"table","data":{"content":" | 平均值 | P50 | P90 | P99 | P999 |
|---|
pulsar-ack-1| 4.32 | 4.41 | 5.35 | 5.86 | 7.65 | pulsar-ack-2| 4.72 | 4.65 | 5.60 | 11.64 | 33.90 | kafka-ack-1| 6.23 | 4.91 | 9.08 | 18.75 | 91.74 | kafka-ack-all| 12.89 | 7.53 | 23.07 | 64.62 | 169.83"}}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","marks":[{"type":"strong"}],"text":"發佈延遲 - 異步本地持久性"}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"使用兩種複製持久性設置（分別爲 ack-1 和 ack-2）和異步本地持久性時，Pulsar 和 Kafka 發佈延遲的差異如圖 13。表 4 爲每種情況下的精確延遲數值。在異步複製持久性（ack-1）下，Kafka 性能更優，但 Pulsar P99 發佈延遲更穩定（低於 5 毫秒），增加複製持久性保證（從 ack-1 到 ack-2）並未影響延遲。在同步複製持久性（ack-2）下，"},{"type":"text","marks":[{"type":"strong"}],"text":"Kafka P99 發佈延遲是 Pulsar 的 2 倍"},{"type":"text","text":"。"}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"image","attrs":{"src":"https:\/\/static001.infoq.cn\/resource\/image\/98\/2b\/98e9918b1b4aee849a6d4d169942792b.png","alt":null,"title":null,"style":[{"key":"width","value":"75%"},{"key":"bordertype","value":"none"}],"href":null,"fromPaste":true,"pastePass":true}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":"center","origin":null},"content":[{"type":"text","text":"圖13 Pulsar 和 Kafka 的發佈延遲(無數據同步)"}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","marks":[{"type":"strong"}],"text":"表 4 Pulsar 和 Kafka 的實際發佈延遲測試結果（無數據同步）"}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"embedcomp","attrs":{"type":"table","data":{"content":" | 平均值 | P50 | P90 | P99 | P999 |
|---|
pulsar-ack-1| 2.81 | 2.79 | 3.8 | 4.21 | 5.45 | pulsar-ack-2| 2.99 | 2.97 | 3.96 | 4.47 | 6.19 | kafka-ack-1| 1.74 | 1.62 | 2.60 | 3.06 | 9.91 | kafka-ack-all| 3.01 | 2.77 | 3.97 | 8.47 | 15.57"}}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"爲了更深入地瞭解發佈延遲如何隨時間變化，我們繪製了 Pulsar 和 Kafka 在不同複製持久性設置下的 P99 發佈延遲圖。如圖 14 所示，Pulsar 的延遲始終保持在較低的水平（低於 5 毫秒），"},{"type":"text","marks":[{"type":"strong"}],"text":"而 Kafka 的延遲始終是 Pulsar 的 2 倍"},{"type":"text","text":"。"}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"image","attrs":{"src":"https:\/\/static001.infoq.cn\/resource\/image\/86\/ae\/86b22dda18605492ce4344370ae7f1ae.png","alt":null,"title":null,"style":[{"key":"width","value":"75%"},{"key":"bordertype","value":"none"}],"href":null,"fromPaste":true,"pastePass":true}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":"center","origin":null},"content":[{"type":"text","text":"圖14 不同複製持久性設置下，Pulsar 和 Kafka 的 P99 發佈延遲"}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","marks":[{"type":"strong"}],"text":"端到端延遲--異步本地持久性"}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"使用兩種複製持久性設置（分別爲 ack-1 和 ack-2）和異步本地持久性時，Pulsar 和 Kafka 端到端延遲的差異如圖 15。表 5 爲每種情況下的精確延遲數值。在所有情況下，Pulsar 的性能都優於 Kafka。Pulsar P99 端到端延遲保持穩定（約 5 毫秒），改變複製持久性設置對其並無影響。對於 ack-1，"},{"type":"text","marks":[{"type":"strong"}],"text":"Kafka 的 P99 端到端延遲比 Pulsar 高；對於 ack-2，Kafka 的 P99 端到端延遲是 Pulsar 的 2 倍"},{"type":"text","text":"。"}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"image","attrs":{"src":"https:\/\/static001.infoq.cn\/resource\/image\/4f\/2f\/4fe43f34203a939fd78b4f0e9a94f22f.png","alt":null,"title":null,"style":[{"key":"width","value":"75%"},{"key":"bordertype","value":"none"}],"href":null,"fromPaste":true,"pastePass":true}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":"center","origin":null},"content":[{"type":"text","text":"圖15 1 個訂閱時，Pulsar 和 Kafka 的端到端延遲（無數據同步）"}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","marks":[{"type":"strong"}],"text":"表 5 1 個訂閱時，Pulsar 和 Kafka 的實際端到端延遲測試結果（無數據同步）"}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"embedcomp","attrs":{"type":"table","data":{"content":" | 平均值 | P50 | P90 | P99 | P999 |
|---|
pulsar-ack-1| 3.96 | 3.99 | 4.90 | 5.33 | 7.93 | pulsar-ack-2| 4.06 | 4.17 | 5.08 | 5.55 | 8.52 | kafka-ack-1| 4.26 | 4.10 | 5.39 | 6.94 | 17.24 | kafka-ack-all| 4.22 | 3.96 | 5.19 | 10.43 | 18.95"}}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","marks":[{"type":"strong"}],"text":"#2 100 個分區, 10 個訂閱"}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"瞭解 Pulsar 和 Kafka 只有一個訂閱時的性能後，我們想知道更改訂閱數量會如何影響發佈延遲和端到端延遲。於是我們把訂閱數量從 1 增加到 10，併爲每個訂閱分配了 2 個消費者。"}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"從表 6 的數據中可以看出，我們的測試結果表明："}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"bulletedlist","content":[{"type":"listitem","attrs":{"listStyle":null},"content":[{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"Pulsar P99 發佈延遲和端到端延遲保持在 5 到 10 毫秒之間。"}]}]},{"type":"listitem","attrs":{"listStyle":null},"content":[{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"增加訂閱數量會對 Kafka 的 P99 發佈和端到端延遲產生巨大影響，導致延遲增加到幾秒。"}]}]}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","marks":[{"type":"strong"}],"text":"表 6 10 個訂閱時，發佈和端到端延遲的測試結果"}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"embedcomp","attrs":{"type":"table","data":{"content":" | 發佈延遲 | 端到端延遲 |
|---|
同步本地持久性| 測試結果 | 測試結果 | 異步本地持久性| 測試結果 | 測試結果"}}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","marks":[{"type":"strong"}],"text":"發佈延遲 - 同步本地持久性"}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"使用兩種複製持久性設置（分別爲 ack-1和 ack-2）和同步本地持久性時，Pulsar 和 Kafka 發佈延遲的差異如圖 16。表 7 爲每種情況下的精確延遲數值。在異步複製持久性（ack-1）下，Pulsar P99 發佈延遲仍然是 Kafka 的 1\/3。但在同步複製持久性（ack-2）下，Pulsar 的發佈延遲是 Kafka 的 1\/160（1 個訂閱時，Pulsar 的發佈延遲是 Kafka 的 1\/5）。"}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"image","attrs":{"src":"https:\/\/static001.infoq.cn\/resource\/image\/16\/d6\/16a6e45bdyyfe063240c59aea7c308d6.png","alt":null,"title":null,"style":[{"key":"width","value":"75%"},{"key":"bordertype","value":"none"}],"href":null,"fromPaste":true,"pastePass":true}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":"center","origin":null},"content":[{"type":"text","text":"圖16 10 個訂閱時，Pulsar 和 Kafka 的發佈延遲（數據同步）"}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","marks":[{"type":"strong"}],"text":"表 7 10 個訂閱時，Pulsar 和 Kafka 的實際發佈延遲測試結果（數據同步）"}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"embedcomp","attrs":{"type":"table","data":{"content":" | 平均值 | P50 | P90 | P99 | P999 |
|---|
pulsar-ack-1| 3.24 | 3.20 | 4.26 | 4.89 | 10.31 | pulsar-ack-2| 3.67 | 3.47 | 4.56 | 9.94 | 31.31 | pafka-ack-1| 3.14 | 2.39 | 4.39 | 15.07 | 61.29 | kafka-ack-all| 290.51 | 176.82 | 724.26 | 1593.46 | 2686.41"}}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","marks":[{"type":"strong"}],"text":"端到端延遲--同步本地持久性"}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"使用兩種複製持久性設置（分別爲 ack-1和 ack-2）和同步本地持久性時，Pulsar 和 Kafka 端到端延遲的差異如圖 17。表 8 爲每種情況下的精確延遲數值。在異步複製持久性（ack-1）下，Pulsar P99 延遲是 Kafka 的 1\/20，在同步複製持久性（ack-2）下是 Kafka 的 1\/110 。"}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"image","attrs":{"src":"https:\/\/static001.infoq.cn\/resource\/image\/ff\/49\/ffba1f49278e1531c4f353c8ae934b49.png","alt":null,"title":null,"style":[{"key":"width","value":"75%"},{"key":"bordertype","value":"none"}],"href":null,"fromPaste":true,"pastePass":true}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":"center","origin":null},"content":[{"type":"text","text":"圖17 10 個訂閱時，Pulsar 和 Kafka 的端到端延遲（數據同步）"}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","marks":[{"type":"strong"}],"text":"表 8 10 個訂閱時，Pulsar 和 Kafka 的實際端到端延遲測試結果(數據同步)"}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"embedcomp","attrs":{"type":"table","data":{"content":" | 平均值 | P50 | P90 | P99 | P999 |
|---|
pulsar-ack-1| 4.79 | 4.83 | 6.03 | 7.12 | 15.36 | pulsar-ack-2| 5.34 | 5.12 | 6.43 | 14.65 | 39.90 | kafka-ack-1| 11.36 | 6.65 | 17.12 | 145.10 | 914.19 | kafka-ack-all| 296.45 | 171.32 | 731.67 | 1599.79 | 2696.63"}}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","marks":[{"type":"strong"}],"text":"發佈延遲 - 異步本地持久性"}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"使用兩種複製持久性設置（分別爲 ack-1 和 ack-2）和異步本地持久性時，Pulsar 和 Kafka 發佈延遲的差異如圖 18。表 9 爲每種情況下的精確延遲數值。Pulsar 的性能顯著優於 Kafka。Pulsar 的平均發佈延遲約爲 3 毫秒，P99 延遲在 5 毫秒以內。Kafka 在異步複製持久性（ack-1）下性能令人滿意，但在同步複製持久性（ack-2）下性能差了很多。在同步複製持久性下，Kafka 的 P99 發佈延遲是 Pulsar 的 270 倍。"}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"image","attrs":{"src":"https:\/\/static001.infoq.cn\/resource\/image\/9a\/2d\/9a63b1e6a8f914c8b23a00dca439872d.png","alt":null,"title":null,"style":[{"key":"width","value":"75%"},{"key":"bordertype","value":"none"}],"href":null,"fromPaste":true,"pastePass":true}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":"center","origin":null},"content":[{"type":"text","text":"圖18 10 個訂閱時，Pulsar 和 Kafka 的發佈延遲（無數據同步）"}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","marks":[{"type":"strong"}],"text":"表 9 10 個訂閱時，Pulsar 和 Kafka 的實際發佈延遲測試結果(無數據同步)"}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"embedcomp","attrs":{"type":"table","data":{"content":" | 平均值 | P50 | P90 | P99 | P999 |
|---|
pulsar-ack-1| 2.86 | 2.82 | 3.86 | 4.46 | 11.18 | pulsar-ack-2| 3.05 | 3.00 | 4.03 | 4.73 | 10.39 | kafka-ack-1| 2.11 | 1.89 | 3.02 | 6.35 | 14.74 | kafka-ack-all| 158.04 | 17.63 | 526.91 | 1281.25 | 1956.71"}}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","marks":[{"type":"strong"}],"text":"端到端延遲--異步本地持久性"}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"使用兩種複製持久性設置（分別爲 ack-1 和 ack-2）和異步本地持久性時，Pulsar 和 Kafka 端到端延遲的差異如圖 19。表 10 爲不同情況下的精確延遲數值。在所有情況下，Pulsar 的性能始終優於 Kafka。Pulsar 的端到端延遲始終保持在 4～7 毫秒之間，改變複製持久性設置對其並無影響。對於 ack-1，Kafka 的 P99 端到端延遲是 Pulsar 的 13 倍；對於ack-2，則是 187 倍。"}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"image","attrs":{"src":"https:\/\/static001.infoq.cn\/resource\/image\/6a\/34\/6aaa289389c4a607949fbfdce178df34.png","alt":null,"title":null,"style":[{"key":"width","value":"75%"},{"key":"bordertype","value":"none"}],"href":null,"fromPaste":true,"pastePass":true}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":"center","origin":null},"content":[{"type":"text","text":"圖19 10 個訂閱時，Pulsar 和 Kafka 的端到端延遲（無數據同步）"}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","marks":[{"type":"strong"}],"text":"表 10 10 個訂閱時，實際端到端延遲測試結果(無數據同步)"}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"embedcomp","attrs":{"type":"table","data":{"content":" | 平均值 | P50 | P90 | P99 | P999 |
|---|
pulsar-ack-1| 4.51 | 4.47 | 5.60 | 6.84 | 15.77 | pulsar-ack-2| 4.61 | 4.61 | 5.76 | 6.94 | 14.21 | kafka-ack-1| 8.01 | 5.90 | 9.38 | 89.80 | 532.68 | kafka-ack-all| 212.77 | 87.72 | 537.85 | 1295.78 | 1971.03"}}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","marks":[{"type":"strong"}],"text":"#3 100, 5000, 8000, 10000 個分區"}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"瞭解更改訂閱數量會如何影響 Pulsar 和 Kafka 的發佈延遲後，我們想更改分區數量，看看會產生什麼效果。於是，我們把分區數量從 100 個累加到 10000 個，觀察發生了什麼變化。"}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"測試數據見表 11，測試結果表明："}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"bulletedlist","content":[{"type":"listitem","attrs":{"listStyle":null},"content":[{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"分區數量增加時，Pulsar P99 發佈延遲穩定在 5 毫秒內。"}]}]},{"type":"listitem","attrs":{"listStyle":null},"content":[{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"Kafka 的 P99 發佈延遲受分區數量累加影響很大，延遲上升到幾秒。"}]}]},{"type":"listitem","attrs":{"listStyle":null},"content":[{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"分區數量超過 5000 後，Kafka 的消費者就跟不上發佈吞吐量了。"}]}]}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","marks":[{"type":"strong"}],"text":"表 11 不同 ack 和不同持久性下的實際發佈延遲測試結果"}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"embedcomp","attrs":{"type":"table","data":{"content":" | Ack = 1 | Ack = 2 |
|---|
| 同步本地持久性 | 測試結果 | 測試結果 | | 異步本地持久性 | 測試結果 | 測試結果"}}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","marks":[{"type":"strong"}],"text":"Ack = 1，同步本地持久性"}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"使用同步本地持久性和異步複製持久性 (ack = 1) 更改分區數量時，Pulsar 和 Kafka 發佈和端到端延遲的差異分別如圖 20 和圖 21。"}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"image","attrs":{"src":"https:\/\/static001.infoq.cn\/resource\/image\/0b\/e1\/0b6ec2e8afc745a2b96ab46f64c187e1.png","alt":null,"title":null,"style":[{"key":"width","value":"75%"},{"key":"bordertype","value":"none"}],"href":null,"fromPaste":true,"pastePass":true}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":"center","origin":null},"content":[{"type":"text","text":"圖20 分區數量不同，ACK=1 時的發佈延遲（數據同步）"}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"image","attrs":{"src":"https:\/\/static001.infoq.cn\/resource\/image\/8f\/11\/8f7669e4363a3476788ced67f9fd8511.png","alt":null,"title":null,"style":[{"key":"width","value":"75%"},{"key":"bordertype","value":"none"}],"href":null,"fromPaste":true,"pastePass":true}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":"center","origin":null},"content":[{"type":"text","text":"圖21 分區數量不同，ACK=1 時的端到端延遲(數據同步)"}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"表 12 顯示了分區數量不同，ACK=1 時的實際發佈延遲測試結果。表 13 顯示了分區數量不同，ACK=1 時的實際端到端延遲測試結果。"}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","marks":[{"type":"strong"}],"text":"表 12 分區數量不同，ACK=1 時的實際發佈延遲測試結果（數據同步）"}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"embedcomp","attrs":{"type":"table","data":{"content":" | 平均值 | P50 | P90 | P99 | P999 |
|---|
kafka-100| 2.54 | 1.99 | 3.78 | 11.37 | 18.71 | kafka-5000| 3.50 | 4.41 | 9.21 | 29.26 | 44.39 | kafka-8000| 8.37 | 5.11 | 20.18 | 40.70 | 69.72 | kafka-10000| 11.14 | 6.57 | 26.81 | 52.24 | 82.07 | pulsar-100| 3.23 | 3.21 | 4.23 | 4.73 | 5.89 | pulsar-5000| 3.35 | 3.30 | 4.34 | 5.03 | 13.96 | pulsar-8000| 3.67 | 3.67 | 4.89 | 5.61 | 16.07 | pulsar-10000| 3.42 | 3.37 | 4.48 | 5.36 | 19.20"}}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","marks":[{"type":"strong"}],"text":"表 13 分區數量不同，ACK=1 時的實際端到端延遲測試結果（數據同步）"}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"embedcomp","attrs":{"type":"table","data":{"content":" | 平均值 | P50 | P90 | P99 | P999 |
|---|
kafka-100| 6.23 | 4.91 | 9.08 | 18.75 | 91.74 | pulsar-100| 4.32 | 4.41 | 5.35 | 5.86 | 7.65 | pulsar-5000| 4.52 | 4.53 | 5.55 | 6.26 | 17.78 | pulsar-8000| 4.89 | 4.99 | 6.11 | 6.86 | 23.83 | pulsar-10000| 4.49 | 4.62 | 5.70 | 6.67 | 27.25"}}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"圖 22 給出了分區數量不同，ACK=1 時 Pulsar 端到端的延遲。圖 23 給出了分區數量不同，ACK=1 時 Kafka 端到端的延遲。"}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"image","attrs":{"src":"https:\/\/static001.infoq.cn\/resource\/image\/dd\/8e\/dd6fc4a2169b5f2c687dd677df95388e.png","alt":null,"title":null,"style":[{"key":"width","value":"75%"},{"key":"bordertype","value":"none"}],"href":null,"fromPaste":true,"pastePass":true}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":"center","origin":null},"content":[{"type":"text","text":"圖22 分區數量不同，ACK=1 時 Pulsar 的端到端延遲（數據同步）"}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"image","attrs":{"src":"https:\/\/static001.infoq.cn\/resource\/image\/f6\/92\/f63a2195bc40c587529aa19b313f5a92.png","alt":null,"title":null,"style":[{"key":"width","value":"75%"},{"key":"bordertype","value":"none"}],"href":null,"fromPaste":true,"pastePass":true}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":"center","origin":null},"content":[{"type":"text","text":"圖23 分區數量不同，ACK=1 時 Kafka 的端到端延遲(數據同步)"}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"從以上圖表可以看出："}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"bulletedlist","content":[{"type":"listitem","attrs":{"listStyle":null},"content":[{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"Pulsar P99 發佈延遲穩定在 5 毫秒左右，更改分區數量對其並無影響。"}]}]},{"type":"listitem","attrs":{"listStyle":null},"content":[{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"Pulsar P99 端到端延遲穩定在 6 毫秒左右，更改分區數量對其並無影響。"}]}]},{"type":"listitem","attrs":{"listStyle":null},"content":[{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"Kafka 的 P99 發佈延遲隨着分區數量增加而逐漸拉長，在 10000 個分區時，其延遲增加了 5 倍（與 100 個分區時相比），比 Pulsar 的延遲高出 10 倍。"}]}]},{"type":"listitem","attrs":{"listStyle":null},"content":[{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"Kafka 的 P99 端到端延遲隨着分區數量增加而逐漸拉長，在 10000 個分區時，其延遲高出 10000 倍（與 100 個分區時相比），增加到 180 秒，是 Pulsar 的 280000倍。"}]}]}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","marks":[{"type":"strong"}],"text":"Ack = 2，同步本地持久性"}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"使用同步本地持久性和同步複製持久性（ack = 2）更改分區數量時，Pulsar 和 Kafka 發佈延遲的差異如圖 24。表 14 爲每種情況下的精切延遲數值。"}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"image","attrs":{"src":"https:\/\/static001.infoq.cn\/resource\/image\/fc\/4e\/fc16d7dbb7df1dec268914a7aa68c34e.png","alt":null,"title":null,"style":[{"key":"width","value":"75%"},{"key":"bordertype","value":"none"}],"href":null,"fromPaste":true,"pastePass":true}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":"center","origin":null},"content":[{"type":"text","text":"圖24 分區數量不同，ack=2\/all 時的發佈延遲（數據同步）"}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","marks":[{"type":"strong"}],"text":"表 14 分區數量不同， ack=2\/all 時的實際發佈延遲測試結果（數據同步）"}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"embedcomp","attrs":{"type":"table","data":{"content":" | 平均值 | P50 | P90 | P99 | P999 |
|---|
kafka-100| 9.84 | 5.71 | 19.7 | 58.83 | 164.20 | kafka-5000| 154.50 | 36.86 | 468.75 | 1259.82 | 2017.15 | kafka-8000| 283.50 | 124.82 | 784.69 | 1742.27 | 2729.79 | kafka-10000| 259.50 | 96.15 | 731.85 | 1718.09 | 2684.28 | pulsar-100| 3.64 | 3.42 | 4.47 | 10.37 | 32.74 | pulsar-5000| 3.84 | 3.57 | 4.69 | 12.81 | 42.83 | pulsar-8000| 4.15 | 3.97 | 5.23 | 11.90 | 42.19"}}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"圖 25 和圖 26 分別顯示了更改分區數量如何影響 Pulsar 和 Kafka 端到端的延遲。"}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"image","attrs":{"src":"https:\/\/static001.infoq.cn\/resource\/image\/59\/45\/59cfbe12e99465c501510b4e5428e945.png","alt":null,"title":null,"style":[{"key":"width","value":"75%"},{"key":"bordertype","value":"none"}],"href":null,"fromPaste":true,"pastePass":true}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":"center","origin":null},"content":[{"type":"text","text":"圖25 分區數量不同，ack=2 時 Pulsar 的端到端延遲（數據同步）"}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"image","attrs":{"src":"https:\/\/static001.infoq.cn\/resource\/image\/f4\/40\/f45a915c2473d850fb3c775939ff1c40.png","alt":null,"title":null,"style":[{"key":"width","value":"75%"},{"key":"bordertype","value":"none"}],"href":null,"fromPaste":true,"pastePass":true}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":"center","origin":null},"content":[{"type":"text","text":"圖26 分區數量不同，ack=2 時 Kafka 的端到端延遲（數據同步）"}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"從以上圖表可以看出："}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"bulletedlist","content":[{"type":"listitem","attrs":{"listStyle":null},"content":[{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"Pulsar P99 發佈延遲穩定在約 10 毫秒，增加分區數量對其並無影響。"}]}]},{"type":"listitem","attrs":{"listStyle":null},"content":[{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"隨着分區數量增加，Kafka 的 P99 發佈延遲逐漸拉長；有 10000 個分區時，其延遲是 100 個分區的 30 倍，增加到 1.7 秒，是 Pulsar 的 126 倍。"}]}]},{"type":"listitem","attrs":{"listStyle":null},"content":[{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"Pulsar P99 端到端延遲穩定在約 10 毫秒，增加分區數量僅對 Pulsar P99 端到端延遲產生輕微影響。但即使有 10000 個分區，延遲仍相對較低，約爲 50 毫秒。"}]}]},{"type":"listitem","attrs":{"listStyle":null},"content":[{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"隨着分區數量增加，Kafka 的 P99 端到端延遲逐漸拉長。有 10000 個分區時，Kafka P99 端到端延遲增加到 200 秒，是 Pulsar 的 14771 倍。"}]}]}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","marks":[{"type":"strong"}],"text":"Ack = 1, Async 本地持久性"}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"使用異步本地持久性和異步複製持久性（ack = 1）更改分區數量時，Pulsar 和 Kafka 發佈延遲的差異如圖 27。表 15 爲不同情況下的精切延遲數值。"}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"image","attrs":{"src":"https:\/\/static001.infoq.cn\/resource\/image\/0c\/d6\/0c02b4488399e82ce8e6f1674f0038d6.png","alt":null,"title":null,"style":[{"key":"width","value":"75%"},{"key":"bordertype","value":"none"}],"href":null,"fromPaste":true,"pastePass":true}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":"center","origin":null},"content":[{"type":"text","text":"圖27 分區數量不同，ack=1 時的發佈延遲（無數據同步）"}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","marks":[{"type":"strong"}],"text":"表 15 分區數量不同，ack=1 時的實際發佈延遲測試結果（無數據同步）"}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"embedcomp","attrs":{"type":"table","data":{"content":" | 平均值 | P50 | P90 | P99 | P999 |
|---|
kafka-100| 1.74 | 1.62 | 2.60 | 3.06 | 9.91 | kafka-5000| 4.13 | 3.05 | 5.20 | 25.32 | 33.86 | kafka-8000| 6.84 | 4.32 | 16.51 | 34.19 | 46.34 | kafka-10000| 8.95 | 5.32 | 22.75 | 41.83 | 59.32 | pulsar-100| 2.86 | 2.79 | 3.8 | 4.21 | 5.45 | pulsar-5000| 2.98 | 2.95 | 3.89 | 4.51 | 7.85 | pulsar-8000| 3.26 | 3.27 | 4.50 | 5.14 | 15.21 | pulsar-10000| 3.06 | 2.97 | 4.15 | 5.23 | 19.47"}}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"圖 28 和圖 29 分別顯示了更改分區數量如何影響 Pulsar 和 Kafka 端到端的延遲。"}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"image","attrs":{"src":"https:\/\/static001.infoq.cn\/resource\/image\/32\/ac\/329f3651ef8bfe94c210c402702101ac.png","alt":null,"title":null,"style":[{"key":"width","value":"75%"},{"key":"bordertype","value":"none"}],"href":null,"fromPaste":true,"pastePass":true}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":"center","origin":null},"content":[{"type":"text","text":"圖28 分區數量不同，ack=1 時，Pulsar 端到端的延遲（無數據同步）"}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"image","attrs":{"src":"https:\/\/static001.infoq.cn\/resource\/image\/96\/98\/96e667ac0efd732ffa0yyc0a0448d598.png","alt":null,"title":null,"style":[{"key":"width","value":"75%"},{"key":"bordertype","value":"none"}],"href":null,"fromPaste":true,"pastePass":true}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":"center","origin":null},"content":[{"type":"text","text":"圖29 分區數量不同，ack=1 時，Kafka 端到端的延遲(無數據同步)"}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"從以上圖表可以看出："}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"bulletedlist","content":[{"type":"listitem","attrs":{"listStyle":null},"content":[{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"Pulsar P99 發佈延遲穩定在約 4～5 毫秒之間，增加分區數量對其並無影響。"}]}]},{"type":"listitem","attrs":{"listStyle":null},"content":[{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"Kafka P99 發佈延遲隨着分區數量增加而逐漸拉長，有 10000 個分區時，延遲增加到 41 毫秒，是有 100 個分區時的 13 倍，是 Pulsar 的 8 倍。"}]}]},{"type":"listitem","attrs":{"listStyle":null},"content":[{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"Pulsar P99 端到端延遲穩定在約 4～6 毫秒之間，增加分區數量僅對 Pulsar P99 端到端延遲產生輕微影響，但即使有 10000 個分區，它也保持在相對較低的水平（24 毫秒以內）。"}]}]},{"type":"listitem","attrs":{"listStyle":null},"content":[{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"隨着分區數量增加，Kafka P99 端到端延遲逐漸拉長。有 10000 個分區時，Kafka P99 端到端延遲高達 180 秒，是 Pulsar 的 34416 倍。"}]}]}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","marks":[{"type":"strong"}],"text":"Ack = 2，異步本地持久性"}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"使用異步本地持久性和同步複製持久性 (ack = 2)更改分區數量時，Pulsar 和 Kafka 發佈延遲的差異如圖 30。表 16 爲每種情況下的精確延遲數值。"}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"image","attrs":{"src":"https:\/\/static001.infoq.cn\/resource\/image\/a7\/e9\/a7a1f13c29956babf847bdb99b6cf0e9.png","alt":null,"title":null,"style":[{"key":"width","value":"75%"},{"key":"bordertype","value":"none"}],"href":null,"fromPaste":true,"pastePass":true}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":"center","origin":null},"content":[{"type":"text","text":"圖30 分區數量不同，ack=all\/2 的發佈延遲（無數據同步）"}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","marks":[{"type":"strong"}],"text":"表 16 分區數量不同，ack=all\/2 的實際發佈延遲測試結果（無數據同步）"}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"embedcomp","attrs":{"type":"table","data":{"content":" | 平均值 | P50 | P90 | P99 | P999 |
|---|
kafka-100| 3.01 | 2.77 | 3.97 | 8.47 | 15.57 | kafka-5000| 19.33 | 10.16 | 40.40 | 121.40 | 336.21 | kafka-8000| 138.19 | 42.52 | 385.86 | 1164.90 | 2008.28 | kafka-10000| 266.66 | 102.55 | 752.95 | 1717.83 | 2797.51 | pulsar-100| 2.99 | 2.97 | 3.96 | 4.47 | 6.19 | | pulsar-5000 | | | | | |
| | | | | | | | |
| | | | | | | |
| | | | | |
| | | | | | | | |
|
| | | | |
| | | | |
| | | | |
| | | | |
| | |
| | | | |
| | | | |
| | | | |
| | | | |
{"type":"doc","content":[{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"爲了更全面地瞭解 Pulsar 和 Kafka，我們“復現”了 Confluent 對 Pulsar 和 Kafka 基準測試。重複這一基準測試的原因有兩個，一是 Confluent 的測試方法存在一些問題；二是 Confluent 的測試範圍和測試場景不夠全面。爲了更準確地對比 Pulsar 和 Kafka，我們在測試中不僅修復了 Confluent 測試中的問題，還擴大了測試範圍，納入更多性能衡量標準，模擬更多實際場景。"}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"和 Confluent 的測試相比，我們的測試主要有三項改進："}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"numberedlist","attrs":{"start":1,"normalizeStart":1},"content":[{"type":"listitem","attrs":{"listStyle":null},"content":[{"type":"paragraph","attrs":{"indent":0,"number":1,"align":null,"origin":null},"content":[{"type":"text","text":"包含 Pulsar 和 Kafka 支持的所有持久性級別。在同等持久性級別下，對比二者的吞吐量和延遲。"}]}]},{"type":"listitem","attrs":{"listStyle":null},"content":[{"type":"paragraph","attrs":{"indent":0,"number":2,"align":null,"origin":null},"content":[{"type":"text","text":"引入影響性能的其他因素和測試條件，如分區數量、訂閱數量、客戶端數量等。"}]}]},{"type":"listitem","attrs":{"listStyle":null},"content":[{"type":"paragraph","attrs":{"indent":0,"number":3,"align":null,"origin":null},"content":[{"type":"text","text":"測試的混合負載同時包含寫入、追趕讀和追尾讀，模擬實際使用場景。"}]}]}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"我們進行了最大吞吐量測試、發佈和端到端延遲測試、追趕讀測試和混合工作負載測試（"},{"type":"link","attrs":{"href":"https:\/\/www.infoq.cn\/article\/HS4gtkQjop4q42z13uEv","title":"xxx","type":null},"content":[{"type":"text","text":"詳見上篇"}]},{"type":"text","text":"）。由於篇幅有限，本篇主要介紹發佈和端到端延遲測試詳情。"}]},{"type":"heading","attrs":{"align":null,"level":2},"content":[{"type":"text","text":"發佈和端到端延遲測試"}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"測試目標：觀測在處理髮布和追尾讀工作負載時 Pulsar 和 Kafka 可實現的最低延遲。"}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"測試設置：通過調整訂閱數量和分區數量，觀測訂閱數量和分區數量對發佈延遲和端到端延遲的影響。"}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"測試策略："}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"bulletedlist","content":[{"type":"listitem","attrs":{"listStyle":null},"content":[{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"將所有消息都複製三次，確保容錯；"}]}]},{"type":"listitem","attrs":{"listStyle":null},"content":[{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"改變 ack 數量，測試在不同持久性保證下，Pulsar 和 Kafka 吞吐量的差異；"}]}]},{"type":"listitem","attrs":{"listStyle":null},"content":[{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"改變訂閱數量（從 1 到 10），分別測試延遲；"}]}]},{"type":"listitem","attrs":{"listStyle":null},"content":[{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"改變分區數量（從 100 到 10000），分別測試延遲；"}]}]},{"type":"listitem","attrs":{"listStyle":null},"content":[{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"消息大小爲 1KB；"}]}]},{"type":"listitem","attrs":{"listStyle":null},"content":[{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"生產者以 200000\/s（約 200MB\/s）的固定速度發送消息，追尾讀消費者在生產者持續發送消息時處理消息。"}]}]}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"各項測試結果如下。"}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","marks":[{"type":"strong"}],"text":"#1 100 個分區, 1 個訂閱"}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"我們從 100 個分區、1 個訂閱開始，在所有不同持久性保證下，對 Pulsar 和 Kafka 能實現的最低延遲進行了基準測試。"}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"測試證明，在所有持久性級別上，Pulsar 的發佈延遲和端到端延遲都比 Kafka 低，是 Kafka 的 1\/5 到 1\/2。"}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","marks":[{"type":"italic"}],"text":"發佈延遲 - 同步本地持久性"}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","text":"使用兩種複製持久性設置（分別爲 ack-1 和 ack-2）和同步本地持久性時，Pulsar 和 Kafka 發佈延遲的差異如圖 10。表 2 爲每種情況下的精確延遲數值。在異步複製持久性（ack-1）下，Pulsar P99 延遲是 Kafka 的 1\/3; 在同步複製持久性（ack-2）下是 Kafka 的 1\/5。"}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"image","attrs":{"src":"https:\/\/static001.infoq.cn\/resource\/image\/c8\/1c\/c8828bd53b91aca9a765f5291d20621c.png","alt":null,"title":null,"style":[{"key":"width","value":"75%"},{"key":"bordertype","value":"none"}],"href":null,"fromPaste":true,"pastePass":true}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":"center","origin":null},"content":[{"type":"text","text":"圖10 Pulsar 和 Kafka 的發佈延遲(數據同步)"}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null},"content":[{"type":"text","marks":[{"type":"strong"}],"text":"表 2 Pulsar 和 Kafka 的實際發佈延遲測試結果（數據同步）"}]},{"type":"paragraph","attrs":{"indent":0,"number":0,"align":null,"origin":null}},{"type":"embedcomp","attrs":{"type":"table","data":{"content":"